feat: web UI dashboard for managing Hermes Agent

Adds an embedded web UI dashboard accessible via `hermes web`. Provides a browser-based interface for: - Monitoring agent status, gateway, and active/recent sessions - Editing config.yaml with a schema-driven form editor - Managing API keys in .env (set, clear, view redacted) Backend: FastAPI server (hermes_cli/web_server.py) with REST endpoints. Frontend: Vite + React + TypeScript + Tailwind v4 SPA (web/ directory). Also adds: - `/reload` slash command for hot-reloading .env variables - `delete_env_value()` and `reload_env()` utilities in config.py - `[web]` optional dependency extra (fastapi + uvicorn) - Web build step in `hermes update` (both git and zip paths) - hermes_cli/web_dist/ to .gitignore and package-data Salvaged from PR #1813 by austinpickett onto current main. Fixes applied during salvage: - Replaced hardcoded CONFIG_SCHEMA with dynamic generation from DEFAULT_CONFIG - Restricted CORS to localhost origins (was allow_origins=[*]) - Dropped _maybe_reload_env from model_tools.py (stale reverts to core dispatch) - Dropped .python-version file - Skipped all stale-branch reverts (pyproject.toml, model_tools.py, etc.)
2026-04-11 01:05:13 -07:00
249 changed files with 9195 additions and 19672 deletions
@@ -89,15 +89,6 @@
 # Optional base URL override:
 # HERMES_QWEN_BASE_URL=https://portal.qwen.ai/v1

-# =============================================================================
-# LLM PROVIDER (Xiaomi MiMo)
-# =============================================================================
-# Xiaomi MiMo models (mimo-v2-pro, mimo-v2-omni, mimo-v2-flash).
-# Get your key at: https://platform.xiaomimimo.com
-# XIAOMI_API_KEY=your_key_here
-# Optional base URL override:
-# XIAOMI_BASE_URL=https://api.xiaomimimo.com/v1
-
 # =============================================================================
 # TOOL API KEYS
 # =============================================================================
@@ -51,6 +51,9 @@ ignored/
 .worktrees/
 environments/benchmarks/evals/

+# Web UI build output
+hermes_cli/web_dist/
+
 # Release script temp files
 .release_notes.md
 mini-swe-agent/
@@ -351,9 +351,8 @@ Cache-breaking forces dramatically higher costs. The ONLY time we alter context

 ### Background Process Notifications (Gateway)

-When `terminal(background=true, notify_on_complete=true)` is used, the gateway runs a watcher that
-detects process completion and triggers a new agent turn. Control verbosity of background process
-messages with `display.background_process_notifications`
+When `terminal(background=true, check_interval=...)` is used, the gateway runs a watcher that
+pushes status updates to the user's chat. Control verbosity with `display.background_process_notifications`
 in config.yaml (or `HERMES_BACKGROUND_NOTIFICATIONS` env var):

 - `all` — running-output updates + final message (default)
@@ -1,44 +1,27 @@
-FROM ghcr.io/astral-sh/uv:0.11.6-python3.13-trixie@sha256:b3c543b6c4f23a5f2df22866bd7857e5d304b67a564f4feab6ac22044dde719b AS uv_source
-FROM tianon/gosu:1.19-trixie@sha256:3b176695959c71e123eb390d427efc665eeb561b1540e82679c15e992006b8b9 AS gosu_source
 FROM debian:13.4

 # Disable Python stdout buffering to ensure logs are printed immediately
 ENV PYTHONUNBUFFERED=1

-# Store Playwright browsers outside the volume mount so the build-time
-# install survives the /opt/data volume overlay at runtime.
-ENV PLAYWRIGHT_BROWSERS_PATH=/opt/hermes/.playwright
-
 # Install system dependencies in one layer, clear APT cache
 RUN apt-get update && \
    apt-get install -y --no-install-recommends \
-        build-essential nodejs npm python3 ripgrep ffmpeg gcc python3-dev libffi-dev procps && \
+        build-essential nodejs npm python3 python3-pip ripgrep ffmpeg gcc python3-dev libffi-dev && \
    rm -rf /var/lib/apt/lists/*

-# Non-root user for runtime; UID can be overridden via HERMES_UID at runtime
-RUN useradd -u 10000 -m -d /opt/data hermes
-
-COPY --chmod=0755 --from=gosu_source /gosu /usr/local/bin/
-COPY --chmod=0755 --from=uv_source /usr/local/bin/uv /usr/local/bin/uvx /usr/local/bin/
-
 COPY . /opt/hermes
 WORKDIR /opt/hermes

-# Install Node dependencies and Playwright as root (--with-deps needs apt)
-RUN npm install --prefer-offline --no-audit && \
+# Install Python and Node dependencies in one layer, no cache
+RUN pip install --no-cache-dir uv --break-system-packages && \
+    uv pip install --system --break-system-packages --no-cache -e ".[all]" && \
+    npm install --prefer-offline --no-audit && \
    npx playwright install --with-deps chromium --only-shell && \
    cd /opt/hermes/scripts/whatsapp-bridge && \
    npm install --prefer-offline --no-audit && \
    npm cache clean --force

-# Hand ownership to hermes user, then install Python deps in a virtualenv
-RUN chown -R hermes:hermes /opt/hermes
-USER hermes
-
-RUN uv venv && \
-    uv pip install --no-cache-dir -e ".[all]"
-
-USER root
+WORKDIR /opt/hermes
 RUN chmod +x /opt/hermes/docker/entrypoint.sh

 ENV HERMES_HOME=/opt/data
@@ -60,8 +60,6 @@ _ANTHROPIC_OUTPUT_LIMITS = {
    "claude-3-opus":       4_096,
    "claude-3-sonnet":     4_096,
    "claude-3-haiku":      4_096,
-    # Third-party Anthropic-compatible providers
-    "minimax":            131_072,
 }

 # For any model not in the table, assume the highest current limit.
@@ -1315,10 +1313,9 @@ def build_anthropic_kwargs(
    # Map reasoning_config to Anthropic's thinking parameter.
    # Claude 4.6 models use adaptive thinking + output_config.effort.
    # Older models use manual thinking with budget_tokens.
-    # MiniMax Anthropic-compat endpoints support thinking (manual mode only,
-    # not adaptive).  Haiku does NOT support extended thinking — skip entirely.
+    # Haiku and MiniMax models do NOT support extended thinking — skip entirely.
    if reasoning_config and isinstance(reasoning_config, dict):
-        if reasoning_config.get("enabled") is not False and "haiku" not in model.lower():
+        if reasoning_config.get("enabled") is not False and "haiku" not in model.lower() and "minimax" not in model.lower():
            effort = str(reasoning_config.get("effort", "medium")).lower()
            budget = THINKING_BUDGET.get(effort, 8000)
            if _supports_adaptive_thinking(model):
@@ -23,13 +23,17 @@ Resolution order for vision/multimodal tasks (auto mode):
  6. Custom endpoint (for local vision models: Qwen-VL, LLaVA, Pixtral, etc.)
  7. None

-Per-task overrides are configured in config.yaml under the ``auxiliary:`` section
-(e.g. ``auxiliary.vision.provider``, ``auxiliary.compression.model``).
+Per-task provider overrides (e.g. AUXILIARY_VISION_PROVIDER,
+CONTEXT_COMPRESSION_PROVIDER) can force a specific provider for each task.
 Default "auto" follows the chains above.

-Legacy env var overrides (AUXILIARY_{TASK}_PROVIDER, AUXILIARY_{TASK}_MODEL,
-AUXILIARY_{TASK}_BASE_URL, etc.) are still read as a backward-compat fallback
-but config.yaml takes priority.  New configuration should always use config.yaml.
+Per-task model overrides (e.g. AUXILIARY_VISION_MODEL,
+AUXILIARY_WEB_EXTRACT_MODEL) let callers use a different model slug
+than the provider's default.
+
+Per-task direct endpoint overrides (e.g. AUXILIARY_VISION_BASE_URL,
+AUXILIARY_VISION_API_KEY) let callers route a specific auxiliary task to a
+custom OpenAI-compatible endpoint without touching the main model settings.

 Payment / credit exhaustion fallback:
  When a resolved provider returns HTTP 402 or a credit-related error,
@@ -55,9 +59,6 @@ from hermes_constants import OPENROUTER_BASE_URL

 logger = logging.getLogger(__name__)

-# Module-level flag: only warn once per process about stale OPENAI_BASE_URL.
-_stale_base_url_warned = False
-
 _PROVIDER_ALIASES = {
    "google": "gemini",
    "google-gemini": "gemini",
@@ -107,14 +108,6 @@ _API_KEY_PROVIDER_AUX_MODELS: Dict[str, str] = {
    "kilocode": "google/gemini-3-flash-preview",
 }

-# Vision-specific model overrides for direct providers.
-# When the user's main provider has a dedicated vision/multimodal model that
-# differs from their main chat model, map it here.  The vision auto-detect
-# "exotic provider" branch checks this before falling back to the main model.
-_PROVIDER_VISION_MODELS: Dict[str, str] = {
-    "xiaomi": "mimo-v2-omni",
-}
-
 # OpenRouter app attribution headers
 _OR_HEADERS = {
    "HTTP-Referer": "https://hermes-agent.nousresearch.com",
@@ -714,9 +707,7 @@ def _resolve_api_key_provider() -> Tuple[Optional[OpenAI], Optional[str]]:
            base_url = _to_openai_base_url(
                _pool_runtime_base_url(entry, pconfig.inference_base_url) or pconfig.inference_base_url
            )
-            model = _API_KEY_PROVIDER_AUX_MODELS.get(provider_id)
-            if model is None:
-                continue  # skip provider if we don't know a valid aux model
+            model = _API_KEY_PROVIDER_AUX_MODELS.get(provider_id, "default")
            logger.debug("Auxiliary text client: %s (%s) via pool", pconfig.name, model)
            extra = {}
            if "api.kimi.com" in base_url.lower():
@@ -735,9 +726,7 @@ def _resolve_api_key_provider() -> Tuple[Optional[OpenAI], Optional[str]]:
        base_url = _to_openai_base_url(
            str(creds.get("base_url", "")).strip().rstrip("/") or pconfig.inference_base_url
        )
-        model = _API_KEY_PROVIDER_AUX_MODELS.get(provider_id)
-        if model is None:
-            continue  # skip provider if we don't know a valid aux model
+        model = _API_KEY_PROVIDER_AUX_MODELS.get(provider_id, "default")
        logger.debug("Auxiliary text client: %s (%s)", pconfig.name, model)
        extra = {}
        if "api.kimi.com" in base_url.lower():
@@ -1086,12 +1075,11 @@ def _is_connection_error(exc: Exception) -> bool:
 def _try_payment_fallback(
    failed_provider: str,
    task: str = None,
-    reason: str = "payment error",
 ) -> Tuple[Optional[Any], Optional[str], str]:
-    """Try alternative providers after a payment/credit or connection error.
+    """Try alternative providers after a payment/credit error.

    Iterates the standard auto-detection chain, skipping the provider that
-    failed.
+    returned a payment error.

    Returns:
        (client, model, provider_label) or (None, None, "") if no fallback.
@@ -1117,15 +1105,15 @@ def _try_payment_fallback(
        client, model = try_fn()
        if client is not None:
            logger.info(
-                "Auxiliary %s: %s on %s — falling back to %s (%s)",
-                task or "call", reason, failed_provider, label, model or "default",
+                "Auxiliary %s: payment error on %s — falling back to %s (%s)",
+                task or "call", failed_provider, label, model or "default",
            )
            return client, model, label
        tried.append(label)

    logger.warning(
-        "Auxiliary %s: %s on %s and no fallback available (tried: %s)",
-        task or "call", reason, failed_provider, ", ".join(tried),
+        "Auxiliary %s: payment error on %s and no fallback available (tried: %s)",
+        task or "call", failed_provider, ", ".join(tried),
    )
    return None, None, ""

@@ -1140,28 +1128,9 @@ def _resolve_auto() -> Tuple[Optional[OpenAI], Optional[str]]:
         provider they already have credentials for — no OpenRouter key needed.
      2. OpenRouter → Nous → custom → Codex → API-key providers (original chain).
    """
-    global auxiliary_is_nous, _stale_base_url_warned
+    global auxiliary_is_nous
    auxiliary_is_nous = False  # Reset — _try_nous() will set True if it wins

-    # ── Warn once if OPENAI_BASE_URL is set but config.yaml uses a named
-    #    provider (not 'custom').  This catches the common "env poisoning"
-    #    scenario where a user switches providers via `hermes model` but the
-    #    old OPENAI_BASE_URL lingers in ~/.hermes/.env. ──
-    if not _stale_base_url_warned:
-        _env_base = os.getenv("OPENAI_BASE_URL", "").strip()
-        _cfg_provider = _read_main_provider()
-        if (_env_base and _cfg_provider
-                and _cfg_provider != "custom"
-                and not _cfg_provider.startswith("custom:")):
-            logger.warning(
-                "OPENAI_BASE_URL is set (%s) but model.provider is '%s'. "
-                "Auxiliary clients may route to the wrong endpoint. "
-                "Run: hermes model to reconfigure, or remove "
-                "OPENAI_BASE_URL from ~/.hermes/.env",
-                _env_base, _cfg_provider,
-            )
-            _stale_base_url_warned = True
-
    # ── Step 1: non-aggregator main provider → use main model directly ──
    main_provider = _read_main_provider()
    main_model = _read_main_model()
@@ -1248,7 +1217,6 @@ def resolve_provider_client(
    raw_codex: bool = False,
    explicit_base_url: str = None,
    explicit_api_key: str = None,
-    api_mode: str = None,
 ) -> Tuple[Optional[Any], Optional[str]]:
    """Central router: given a provider name and optional model, return a
    configured client with the correct auth, base URL, and API format.
@@ -1272,10 +1240,6 @@ def resolve_provider_client(
            the main agent loop).
        explicit_base_url: Optional direct OpenAI-compatible endpoint.
        explicit_api_key: Optional API key paired with explicit_base_url.
-        api_mode: API mode override.  One of "chat_completions",
-            "codex_responses", or None (auto-detect).  When set to
-            "codex_responses", the client is wrapped in
-            CodexAuxiliaryClient to route through the Responses API.

    Returns:
        (client, resolved_model) or (None, None) if auth is unavailable.
@@ -1283,40 +1247,6 @@ def resolve_provider_client(
    # Normalise aliases
    provider = _normalize_aux_provider(provider)

-    def _needs_codex_wrap(client_obj, base_url_str: str, model_str: str) -> bool:
-        """Decide if a plain OpenAI client should be wrapped for Responses API.
-
-        Returns True when api_mode is explicitly "codex_responses", or when
-        auto-detection (api.openai.com + codex-family model) suggests it.
-        Already-wrapped clients (CodexAuxiliaryClient) are skipped.
-        """
-        if isinstance(client_obj, CodexAuxiliaryClient):
-            return False
-        if raw_codex:
-            return False
-        if api_mode == "codex_responses":
-            return True
-        # Auto-detect: api.openai.com + codex model name pattern
-        if api_mode and api_mode != "codex_responses":
-            return False  # explicit non-codex mode
-        normalized_base = (base_url_str or "").strip().lower()
-        if "api.openai.com" in normalized_base and "openrouter" not in normalized_base:
-            model_lower = (model_str or "").lower()
-            if "codex" in model_lower:
-                return True
-        return False
-
-    def _wrap_if_needed(client_obj, final_model_str: str, base_url_str: str = ""):
-        """Wrap a plain OpenAI client in CodexAuxiliaryClient if Responses API is needed."""
-        if _needs_codex_wrap(client_obj, base_url_str, final_model_str):
-            logger.debug(
-                "resolve_provider_client: wrapping client in CodexAuxiliaryClient "
-                "(api_mode=%s, model=%s, base_url=%s)",
-                api_mode or "auto-detected", final_model_str,
-                base_url_str[:60] if base_url_str else "")
-            return CodexAuxiliaryClient(client_obj, final_model_str)
-        return client_obj
-
    # ── Auto: try all providers in priority order ────────────────────
    if provider == "auto":
        client, resolved = _resolve_auto()
@@ -1406,7 +1336,6 @@ def resolve_provider_client(
                from hermes_cli.models import copilot_default_headers
                extra["default_headers"] = copilot_default_headers()
            client = OpenAI(api_key=custom_key, base_url=custom_base, **extra)
-            client = _wrap_if_needed(client, final_model, custom_base)
            return (_to_async_client(client, final_model) if async_mode
                    else (client, final_model))
        # Try custom first, then codex, then API-key providers
@@ -1415,8 +1344,6 @@ def resolve_provider_client(
            client, default = try_fn()
            if client is not None:
                final_model = _normalize_resolved_model(model or default, provider)
-                _cbase = str(getattr(client, "base_url", "") or "")
-                client = _wrap_if_needed(client, final_model, _cbase)
                return (_to_async_client(client, final_model) if async_mode
                        else (client, final_model))
        logger.warning("resolve_provider_client: custom/main requested "
@@ -1436,7 +1363,6 @@ def resolve_provider_client(
                    provider,
                )
                client = OpenAI(api_key=custom_key, base_url=custom_base)
-                client = _wrap_if_needed(client, final_model, custom_base)
                logger.debug(
                    "resolve_provider_client: named custom provider %r (%s)",
                    provider, final_model)
@@ -1516,11 +1442,6 @@ def resolve_provider_client(
            except ImportError:
                pass

-        # Honor api_mode for any API-key provider (e.g. direct OpenAI with
-        # codex-family models).  The copilot-specific wrapping above handles
-        # copilot; this covers the general case (#6800).
-        client = _wrap_if_needed(client, final_model, base_url)
-
        logger.debug("resolve_provider_client: %s (%s)", provider, final_model)
        return (_to_async_client(client, final_model) if async_mode
                else (client, final_model))
@@ -1553,13 +1474,12 @@ def get_text_auxiliary_client(task: str = "") -> Tuple[Optional[OpenAI], Optiona
    Callers may override the returned model with a per-task env var
    (e.g. CONTEXT_COMPRESSION_MODEL, AUXILIARY_WEB_EXTRACT_MODEL).
    """
-    provider, model, base_url, api_key, api_mode = _resolve_task_provider_model(task or None)
+    provider, model, base_url, api_key = _resolve_task_provider_model(task or None)
    return resolve_provider_client(
        provider,
        model=model,
        explicit_base_url=base_url,
        explicit_api_key=api_key,
-        api_mode=api_mode,
    )


@@ -1570,14 +1490,13 @@ def get_async_text_auxiliary_client(task: str = ""):
    (AsyncCodexAuxiliaryClient, model) which wraps the Responses API.
    Returns (None, None) when no provider is available.
    """
-    provider, model, base_url, api_key, api_mode = _resolve_task_provider_model(task or None)
+    provider, model, base_url, api_key = _resolve_task_provider_model(task or None)
    return resolve_provider_client(
        provider,
        model=model,
        async_mode=True,
        explicit_base_url=base_url,
        explicit_api_key=api_key,
-        api_mode=api_mode,
    )


@@ -1650,7 +1569,7 @@ def resolve_vision_provider_client(
    backends, so users can intentionally force experimental providers. Auto mode
    stays conservative and only tries vision backends known to work today.
    """
-    requested, resolved_model, resolved_base_url, resolved_api_key, resolved_api_mode = _resolve_task_provider_model(
+    requested, resolved_model, resolved_base_url, resolved_api_key = _resolve_task_provider_model(
        "vision", provider, model, base_url, api_key
    )
    requested = _normalize_vision_provider(requested)
@@ -1691,18 +1610,16 @@ def resolve_vision_provider_client(
                if sync_client is not None:
                    return _finalize(main_provider, sync_client, default_model)
            else:
-                # Exotic provider (DeepSeek, Alibaba, Xiaomi, named custom, etc.)
-                # Use provider-specific vision model if available, otherwise main model.
-                vision_model = _PROVIDER_VISION_MODELS.get(main_provider, main_model)
+                # Exotic provider (DeepSeek, Alibaba, named custom, etc.)
                rpc_client, rpc_model = resolve_provider_client(
-                    main_provider, vision_model)
+                    main_provider, main_model)
                if rpc_client is not None:
                    logger.info(
                        "Vision auto-detect: using active provider %s (%s)",
-                        main_provider, rpc_model or vision_model,
+                        main_provider, rpc_model or main_model,
                    )
                    return _finalize(
-                        main_provider, rpc_client, rpc_model or vision_model)
+                        main_provider, rpc_client, rpc_model or main_model)

        # Fall back through aggregators.
        for candidate in _VISION_AUTO_PROVIDER_ORDER:
@@ -1868,30 +1785,12 @@ def cleanup_stale_async_clients() -> None:
            del _client_cache[key]


-def _is_openrouter_client(client: Any) -> bool:
-    for obj in (client, getattr(client, "_client", None), getattr(client, "client", None)):
-        if obj and "openrouter" in str(getattr(obj, "base_url", "") or "").lower():
-            return True
-    return False
-
-
-def _compat_model(client: Any, model: Optional[str], cached_default: Optional[str]) -> Optional[str]:
-    """Drop OpenRouter-format model slugs (with '/') for non-OpenRouter clients.
-
-    Mirrors the guard in resolve_provider_client() which is skipped on cache hits.
-    """
-    if model and "/" in model and not _is_openrouter_client(client):
-        return cached_default
-    return model or cached_default
-
-
 def _get_cached_client(
    provider: str,
    model: str = None,
    async_mode: bool = False,
    base_url: str = None,
    api_key: str = None,
-    api_mode: str = None,
 ) -> Tuple[Optional[Any], Optional[str]]:
    """Get or create a cached client for the given provider.

@@ -1915,7 +1814,7 @@ def _get_cached_client(
            loop_id = id(current_loop)
        except RuntimeError:
            pass
-    cache_key = (provider, async_mode, base_url or "", api_key or "", api_mode or "", loop_id)
+    cache_key = (provider, async_mode, base_url or "", api_key or "", loop_id)
    with _client_cache_lock:
        if cache_key in _client_cache:
            cached_client, cached_default, cached_loop = _client_cache[cache_key]
@@ -1927,11 +1826,9 @@ def _get_cached_client(
                    _force_close_async_httpx(cached_client)
                    del _client_cache[cache_key]
                else:
-                    effective = _compat_model(cached_client, model, cached_default)
-                    return cached_client, effective
+                    return cached_client, model or cached_default
            else:
-                effective = _compat_model(cached_client, model, cached_default)
-                return cached_client, effective
+                return cached_client, model or cached_default
    # Build outside the lock
    client, default_model = resolve_provider_client(
        provider,
@@ -1939,7 +1836,6 @@ def _get_cached_client(
        async_mode,
        explicit_base_url=base_url,
        explicit_api_key=api_key,
-        api_mode=api_mode,
    )
    if client is not None:
        # For async clients, remember which loop they were created on so we
@@ -1959,26 +1855,24 @@ def _resolve_task_provider_model(
    model: str = None,
    base_url: str = None,
    api_key: str = None,
-) -> Tuple[str, Optional[str], Optional[str], Optional[str], Optional[str]]:
+) -> Tuple[str, Optional[str], Optional[str], Optional[str]]:
    """Determine provider + model for a call.

    Priority:
      1. Explicit provider/model/base_url/api_key args (always win)
-      2. Config file (auxiliary.{task}.* or compression.*)
-      3. Env var overrides (backward-compat: AUXILIARY_{TASK}_*, CONTEXT_{TASK}_*)
+      2. Env var overrides (AUXILIARY_{TASK}_*, CONTEXT_{TASK}_*)
+      3. Config file (auxiliary.{task}.* or compression.*)
      4. "auto" (full auto-detection chain)

-    Returns (provider, model, base_url, api_key, api_mode) where model may
-    be None (use provider default). When base_url is set, provider is forced
-    to "custom" and the task uses that direct endpoint. api_mode is one of
-    "chat_completions", "codex_responses", or None (auto-detect).
+    Returns (provider, model, base_url, api_key) where model may be None
+    (use provider default). When base_url is set, provider is forced to
+    "custom" and the task uses that direct endpoint.
    """
    config = {}
    cfg_provider = None
    cfg_model = None
    cfg_base_url = None
    cfg_api_key = None
-    cfg_api_mode = None

    if task:
        try:
@@ -1995,7 +1889,6 @@ def _resolve_task_provider_model(
        cfg_model = str(task_config.get("model", "")).strip() or None
        cfg_base_url = str(task_config.get("base_url", "")).strip() or None
        cfg_api_key = str(task_config.get("api_key", "")).strip() or None
-        cfg_api_mode = str(task_config.get("api_mode", "")).strip() or None

        # Backwards compat: compression section has its own keys.
        # The auxiliary.compression defaults to provider="auto", so treat
@@ -2008,38 +1901,31 @@ def _resolve_task_provider_model(
                _sbu = comp.get("summary_base_url") or ""
                cfg_base_url = cfg_base_url or _sbu.strip() or None

-    # Env vars are backward-compat fallback only — config.yaml is primary.
    env_model = _get_auxiliary_env_override(task, "MODEL") if task else None
-    env_api_mode = _get_auxiliary_env_override(task, "API_MODE") if task else None
-    resolved_model = model or cfg_model or env_model
-    resolved_api_mode = cfg_api_mode or env_api_mode
+    resolved_model = model or env_model or cfg_model

    if base_url:
-        return "custom", resolved_model, base_url, api_key, resolved_api_mode
+        return "custom", resolved_model, base_url, api_key
    if provider:
-        return provider, resolved_model, base_url, api_key, resolved_api_mode
+        return provider, resolved_model, base_url, api_key

    if task:
-        # Config.yaml is the primary source for per-task overrides.
-        if cfg_base_url:
-            return "custom", resolved_model, cfg_base_url, cfg_api_key, resolved_api_mode
-        if cfg_provider and cfg_provider != "auto":
-            return cfg_provider, resolved_model, None, None, resolved_api_mode
-
-        # Env vars are backward-compat fallback for users who haven't
-        # migrated to config.yaml yet.
        env_base_url = _get_auxiliary_env_override(task, "BASE_URL")
        env_api_key = _get_auxiliary_env_override(task, "API_KEY")
        if env_base_url:
-            return "custom", resolved_model, env_base_url, env_api_key, resolved_api_mode
+            return "custom", resolved_model, env_base_url, env_api_key or cfg_api_key

        env_provider = _get_auxiliary_provider(task)
        if env_provider != "auto":
-            return env_provider, resolved_model, None, None, resolved_api_mode
+            return env_provider, resolved_model, None, None

-        return "auto", resolved_model, None, None, resolved_api_mode
+        if cfg_base_url:
+            return "custom", resolved_model, cfg_base_url, cfg_api_key
+        if cfg_provider and cfg_provider != "auto":
+            return cfg_provider, resolved_model, None, None
+        return "auto", resolved_model, None, None

-    return "auto", resolved_model, None, None, resolved_api_mode
+    return "auto", resolved_model, None, None


 _DEFAULT_AUX_TIMEOUT = 30.0
@@ -2111,37 +1997,6 @@ def _build_call_kwargs(
    return kwargs


-def _validate_llm_response(response: Any, task: str = None) -> Any:
-    """Validate that an LLM response has the expected .choices[0].message shape.
-
-    Fails fast with a clear error instead of letting malformed payloads
-    propagate to downstream consumers where they crash with misleading
-    AttributeError (e.g. "'str' object has no attribute 'choices'").
-
-    See #7264.
-    """
-    if response is None:
-        raise RuntimeError(
-            f"Auxiliary {task or 'call'}: LLM returned None response"
-        )
-    # Allow SimpleNamespace responses from adapters (CodexAuxiliaryClient,
-    # AnthropicAuxiliaryClient) — they have .choices[0].message.
-    try:
-        choices = response.choices
-        if not choices or not hasattr(choices[0], "message"):
-            raise AttributeError("missing choices[0].message")
-    except (AttributeError, TypeError, IndexError) as exc:
-        response_type = type(response).__name__
-        response_preview = str(response)[:120]
-        raise RuntimeError(
-            f"Auxiliary {task or 'call'}: LLM returned invalid response "
-            f"(type={response_type}): {response_preview!r}. "
-            f"Expected object with .choices[0].message — check provider "
-            f"adapter or custom endpoint compatibility."
-        ) from exc
-    return response
-
-
 def call_llm(
    task: str = None,
    *,
@@ -2180,7 +2035,7 @@ def call_llm(
    Raises:
        RuntimeError: If no provider is configured.
    """
-    resolved_provider, resolved_model, resolved_base_url, resolved_api_key, resolved_api_mode = _resolve_task_provider_model(
+    resolved_provider, resolved_model, resolved_base_url, resolved_api_key = _resolve_task_provider_model(
        task, provider, model, base_url, api_key)

    if task == "vision":
@@ -2213,7 +2068,6 @@ def call_llm(
            resolved_model,
            base_url=resolved_base_url,
            api_key=resolved_api_key,
-            api_mode=resolved_api_mode,
        )
        if client is None:
            # When the user explicitly chose a non-OpenRouter provider but no
@@ -2257,20 +2111,18 @@ def call_llm(

    # Handle max_tokens vs max_completion_tokens retry, then payment fallback.
    try:
-        return _validate_llm_response(
-            client.chat.completions.create(**kwargs), task)
+        return client.chat.completions.create(**kwargs)
    except Exception as first_err:
        err_str = str(first_err)
        if "max_tokens" in err_str or "unsupported_parameter" in err_str:
            kwargs.pop("max_tokens", None)
            kwargs["max_completion_tokens"] = max_tokens
            try:
-                return _validate_llm_response(
-                    client.chat.completions.create(**kwargs), task)
+                return client.chat.completions.create(**kwargs)
            except Exception as retry_err:
-                # If the max_tokens retry also hits a payment or connection
-                # error, fall through to the fallback chain below.
-                if not (_is_payment_error(retry_err) or _is_connection_error(retry_err)):
+                # If the max_tokens retry also hits a payment error,
+                # fall through to the payment fallback below.
+                if not _is_payment_error(retry_err):
                    raise
                first_err = retry_err

@@ -2287,24 +2139,19 @@ def call_llm(
        # and providers the user never configured that got picked up by
        # the auto-detection chain.
        should_fallback = _is_payment_error(first_err) or _is_connection_error(first_err)
-        # Only try alternative providers when the user didn't explicitly
-        # configure this task's provider.  Explicit provider = hard constraint;
-        # auto (the default) = best-effort fallback chain.  (#7559)
-        is_auto = resolved_provider in ("auto", "", None)
-        if should_fallback and is_auto:
+        if should_fallback:
            reason = "payment error" if _is_payment_error(first_err) else "connection error"
            logger.info("Auxiliary %s: %s on %s (%s), trying fallback",
                        task or "call", reason, resolved_provider, first_err)
            fb_client, fb_model, fb_label = _try_payment_fallback(
-                resolved_provider, task, reason=reason)
+                resolved_provider, task)
            if fb_client is not None:
                fb_kwargs = _build_call_kwargs(
                    fb_label, fb_model, messages,
                    temperature=temperature, max_tokens=max_tokens,
                    tools=tools, timeout=effective_timeout,
                    extra_body=extra_body)
-                return _validate_llm_response(
-                    fb_client.chat.completions.create(**fb_kwargs), task)
+                return fb_client.chat.completions.create(**fb_kwargs)
        raise


@@ -2382,7 +2229,7 @@ async def async_call_llm(

    Same as call_llm() but async. See call_llm() for full documentation.
    """
-    resolved_provider, resolved_model, resolved_base_url, resolved_api_key, resolved_api_mode = _resolve_task_provider_model(
+    resolved_provider, resolved_model, resolved_base_url, resolved_api_key = _resolve_task_provider_model(
        task, provider, model, base_url, api_key)

    if task == "vision":
@@ -2416,7 +2263,6 @@ async def async_call_llm(
            async_mode=True,
            base_url=resolved_base_url,
            api_key=resolved_api_key,
-            api_mode=resolved_api_mode,
        )
        if client is None:
            _explicit = (resolved_provider or "").strip().lower()
@@ -2427,9 +2273,11 @@ async def async_call_llm(
                    f"variable, or switch to a different provider with `hermes model`."
                )
            if not resolved_base_url:
-                logger.info("Auxiliary %s: provider %s unavailable, trying auto-detection chain",
-                            task or "call", resolved_provider)
-                client, final_model = _get_cached_client("auto", async_mode=True)
+                logger.warning("Provider %s unavailable, falling back to openrouter",
+                               resolved_provider)
+                client, final_model = _get_cached_client(
+                    "openrouter", resolved_model or _OPENROUTER_MODEL,
+                    async_mode=True)
        if client is None:
            raise RuntimeError(
                f"No LLM provider configured for task={task} provider={resolved_provider}. "
@@ -2444,42 +2292,11 @@ async def async_call_llm(
        base_url=resolved_base_url)

    try:
-        return _validate_llm_response(
-            await client.chat.completions.create(**kwargs), task)
+        return await client.chat.completions.create(**kwargs)
    except Exception as first_err:
        err_str = str(first_err)
        if "max_tokens" in err_str or "unsupported_parameter" in err_str:
            kwargs.pop("max_tokens", None)
            kwargs["max_completion_tokens"] = max_tokens
-            try:
-                return _validate_llm_response(
-                    await client.chat.completions.create(**kwargs), task)
-            except Exception as retry_err:
-                # If the max_tokens retry also hits a payment or connection
-                # error, fall through to the fallback chain below.
-                if not (_is_payment_error(retry_err) or _is_connection_error(retry_err)):
-                    raise
-                first_err = retry_err
-
-        # ── Payment / connection fallback (mirrors sync call_llm) ─────
-        should_fallback = _is_payment_error(first_err) or _is_connection_error(first_err)
-        is_auto = resolved_provider in ("auto", "", None)
-        if should_fallback and is_auto:
-            reason = "payment error" if _is_payment_error(first_err) else "connection error"
-            logger.info("Auxiliary %s (async): %s on %s (%s), trying fallback",
-                        task or "call", reason, resolved_provider, first_err)
-            fb_client, fb_model, fb_label = _try_payment_fallback(
-                resolved_provider, task, reason=reason)
-            if fb_client is not None:
-                fb_kwargs = _build_call_kwargs(
-                    fb_label, fb_model, messages,
-                    temperature=temperature, max_tokens=max_tokens,
-                    tools=tools, timeout=effective_timeout,
-                    extra_body=extra_body)
-                # Convert sync fallback client to async
-                async_fb, async_fb_model = _to_async_client(fb_client, fb_model or "")
-                if async_fb_model and async_fb_model != fb_kwargs.get("model"):
-                    fb_kwargs["model"] = async_fb_model
-                return _validate_llm_response(
-                    await async_fb.chat.completions.create(**fb_kwargs), task)
+            return await client.chat.completions.create(**kwargs)
        raise
@@ -4,12 +4,8 @@ Self-contained class with its own OpenAI client for summarization.
 Uses auxiliary model (cheap/fast) to summarize middle turns while
 protecting head and tail context.

-Improvements over v2:
-  - Structured summary template with Resolved/Pending question tracking
-  - Summarizer preamble: "Do not respond to any questions" (from OpenCode)
-  - Handoff framing: "different assistant" (from Codex) to create separation
-  - "Remaining Work" replaces "Next Steps" to avoid reading as active instructions
-  - Clear separator when summary merges into tail message
+Improvements over v1:
+  - Structured summary template (Goal, Progress, Decisions, Files, Next Steps)
  - Iterative summary updates (preserves info across multiple compactions)
  - Token-budget tail protection instead of fixed message count
  - Tool output pruning before LLM summarization (cheap pre-pass)
@@ -24,7 +20,6 @@ from typing import Any, Dict, List, Optional
 from agent.auxiliary_client import call_llm
 from agent.context_engine import ContextEngine
 from agent.model_metadata import (
-    MINIMUM_CONTEXT_LENGTH,
    get_model_context_length,
    estimate_messages_tokens_rough,
 )
@@ -32,13 +27,12 @@ from agent.model_metadata import (
 logger = logging.getLogger(__name__)

 SUMMARY_PREFIX = (
-    "[CONTEXT COMPACTION — REFERENCE ONLY] Earlier turns were compacted "
-    "into the summary below. This is a handoff from a previous context "
-    "window — treat it as background reference, NOT as active instructions. "
-    "Do NOT answer questions or fulfill requests mentioned in this summary; "
-    "they were already addressed. Respond ONLY to the latest user message "
-    "that appears AFTER this summary. The current session state (files, "
-    "config, etc.) may reflect work described here — avoid repeating it:"
+    "[CONTEXT COMPACTION] Earlier turns in this conversation were compacted "
+    "to save context space. The summary below describes work that was "
+    "already completed, and the current session state may still reflect "
+    "that work (for example, files may already be changed). Use the summary "
+    "and the current state to continue from where things left off, and "
+    "avoid repeating work:"
 )
 LEGACY_SUMMARY_PREFIX = "[CONTEXT SUMMARY]:"

@@ -93,10 +87,7 @@ class ContextCompressor(ContextEngine):
        self.api_key = api_key
        self.provider = provider
        self.context_length = context_length
-        self.threshold_tokens = max(
-            int(context_length * self.threshold_percent),
-            MINIMUM_CONTEXT_LENGTH,
-        )
+        self.threshold_tokens = int(context_length * self.threshold_percent)

    def __init__(
        self,
@@ -127,14 +118,7 @@ class ContextCompressor(ContextEngine):
            config_context_length=config_context_length,
            provider=provider,
        )
-        # Floor: never compress below MINIMUM_CONTEXT_LENGTH tokens even if
-        # the percentage would suggest a lower value.  This prevents premature
-        # compression on large-context models at 50% while keeping the % sane
-        # for models right at the minimum.
-        self.threshold_tokens = max(
-            int(self.context_length * threshold_percent),
-            MINIMUM_CONTEXT_LENGTH,
-        )
+        self.threshold_tokens = int(self.context_length * threshold_percent)
        self.compression_count = 0

        # Derive token budgets: ratio is relative to the threshold, not total context
@@ -311,20 +295,13 @@ class ContextCompressor(ContextEngine):

        return "\n\n".join(parts)

-    def _generate_summary(self, turns_to_summarize: List[Dict[str, Any]], focus_topic: str = None) -> Optional[str]:
+    def _generate_summary(self, turns_to_summarize: List[Dict[str, Any]]) -> Optional[str]:
        """Generate a structured summary of conversation turns.

-        Uses a structured template (Goal, Progress, Decisions, Resolved/Pending
-        Questions, Files, Remaining Work) with explicit preamble telling the
-        summarizer not to answer questions.  When a previous summary exists,
+        Uses a structured template (Goal, Progress, Decisions, Files, Next Steps)
+        inspired by Pi-mono and OpenCode. When a previous summary exists,
        generates an iterative update instead of summarizing from scratch.

-        Args:
-            focus_topic: Optional focus string for guided compression.  When
-                provided, the summariser prioritises preserving information
-                related to this topic and is more aggressive about compressing
-                everything else.  Inspired by Claude Code's ``/compact``.
-
        Returns None if all attempts fail — the caller should drop
        the middle turns without a summary rather than inject a useless
        placeholder.
@@ -340,27 +317,60 @@ class ContextCompressor(ContextEngine):
        summary_budget = self._compute_summary_budget(turns_to_summarize)
        content_to_summarize = self._serialize_for_summary(turns_to_summarize)

-        # Preamble shared by both first-compaction and iterative-update prompts.
-        # Inspired by OpenCode's "do not respond to any questions" instruction
-        # and Codex's "another language model" framing.
-        _summarizer_preamble = (
-            "You are a summarization agent creating a context checkpoint. "
-            "Your output will be injected as reference material for a DIFFERENT "
-            "assistant that continues the conversation. "
-            "Do NOT respond to any questions or requests in the conversation — "
-            "only output the structured summary. "
-            "Do NOT include any preamble, greeting, or prefix."
-        )
+        if self._previous_summary:
+            # Iterative update: preserve existing info, add new progress
+            prompt = f"""You are updating a context compaction summary. A previous compaction produced the summary below. New conversation turns have occurred since then and need to be incorporated.

-        # Shared structured template (used by both paths).
-        # Key changes vs v1:
-        #   - "Pending User Asks" section (from Claude Code) explicitly tracks
-        #     unanswered questions so the model knows what's resolved vs open
-        #   - "Remaining Work" replaces "Next Steps" to avoid reading as active
-        #     instructions
-        #   - "Resolved Questions" makes it clear which questions were already
-        #     answered (prevents model from re-answering them)
-        _template_sections = f"""## Goal
+PREVIOUS SUMMARY:
+{self._previous_summary}
+
+NEW TURNS TO INCORPORATE:
+{content_to_summarize}
+
+Update the summary using this exact structure. PRESERVE all existing information that is still relevant. ADD new progress. Move items from "In Progress" to "Done" when completed. Remove information only if it is clearly obsolete.
+
+## Goal
+[What the user is trying to accomplish — preserve from previous summary, update if goal evolved]
+
+## Constraints & Preferences
+[User preferences, coding style, constraints, important decisions — accumulate across compactions]
+
+## Progress
+### Done
+[Completed work — include specific file paths, commands run, results obtained]
+### In Progress
+[Work currently underway]
+### Blocked
+[Any blockers or issues encountered]
+
+## Key Decisions
+[Important technical decisions and why they were made]
+
+## Relevant Files
+[Files read, modified, or created — with brief note on each. Accumulate across compactions.]
+
+## Next Steps
+[What needs to happen next to continue the work]
+
+## Critical Context
+[Any specific values, error messages, configuration details, or data that would be lost without explicit preservation]
+
+## Tools & Patterns
+[Which tools were used, how they were used effectively, and any tool-specific discoveries. Accumulate across compactions.]
+
+Target ~{summary_budget} tokens. Be specific — include file paths, command outputs, error messages, and concrete values rather than vague descriptions.
+
+Write only the summary body. Do not include any preamble or prefix."""
+        else:
+            # First compaction: summarize from scratch
+            prompt = f"""Create a structured handoff summary for a later assistant that will continue this conversation after earlier turns are compacted.
+
+TURNS TO SUMMARIZE:
+{content_to_summarize}
+
+Use this exact structure:
+
+## Goal
 [What the user is trying to accomplish]

 ## Constraints & Preferences
@@ -377,64 +387,22 @@ class ContextCompressor(ContextEngine):
 ## Key Decisions
 [Important technical decisions and why they were made]

-## Resolved Questions
-[Questions the user asked that were ALREADY answered — include the answer so the next assistant does not re-answer them]
-
-## Pending User Asks
-[Questions or requests from the user that have NOT yet been answered or fulfilled. If none, write "None."]
-
 ## Relevant Files
 [Files read, modified, or created — with brief note on each]

-## Remaining Work
-[What remains to be done — framed as context, not instructions]
+## Next Steps
+[What needs to happen next to continue the work]

 ## Critical Context
 [Any specific values, error messages, configuration details, or data that would be lost without explicit preservation]

 ## Tools & Patterns
-[Which tools were used, how they were used effectively, and any tool-specific discoveries]
+[Which tools were used, how they were used effectively, and any tool-specific discoveries (e.g., preferred flags, working invocations, successful command patterns)]

-Target ~{summary_budget} tokens. Be specific — include file paths, command outputs, error messages, and concrete values rather than vague descriptions.
+Target ~{summary_budget} tokens. Be specific — include file paths, command outputs, error messages, and concrete values rather than vague descriptions. The goal is to prevent the next assistant from repeating work or losing important details.

 Write only the summary body. Do not include any preamble or prefix."""

-        if self._previous_summary:
-            # Iterative update: preserve existing info, add new progress
-            prompt = f"""{_summarizer_preamble}
-
-You are updating a context compaction summary. A previous compaction produced the summary below. New conversation turns have occurred since then and need to be incorporated.
-
-PREVIOUS SUMMARY:
-{self._previous_summary}
-
-NEW TURNS TO INCORPORATE:
-{content_to_summarize}
-
-Update the summary using this exact structure. PRESERVE all existing information that is still relevant. ADD new progress. Move items from "In Progress" to "Done" when completed. Move answered questions to "Resolved Questions". Remove information only if it is clearly obsolete.
-
-{_template_sections}"""
-        else:
-            # First compaction: summarize from scratch
-            prompt = f"""{_summarizer_preamble}
-
-Create a structured handoff summary for a different assistant that will continue this conversation after earlier turns are compacted. The next assistant should be able to understand what happened without re-reading the original turns.
-
-TURNS TO SUMMARIZE:
-{content_to_summarize}
-
-Use this exact structure:
-
-{_template_sections}"""
-
-        # Inject focus topic guidance when the user provides one via /compress <focus>.
-        # This goes at the end of the prompt so it takes precedence.
-        if focus_topic:
-            prompt += f"""
-
-FOCUS TOPIC: "{focus_topic}"
-The user has requested that this compaction PRIORITISE preserving all information related to the focus topic above. For content related to "{focus_topic}", include full detail — exact values, file paths, command outputs, error messages, and decisions. For content NOT related to the focus topic, summarise more aggressively (brief one-liners or omit if truly irrelevant). The focus topic sections should receive roughly 60-70% of the summary token budget."""
-
        try:
            call_kwargs = {
                "task": "compression",
@@ -652,7 +620,7 @@ The user has requested that this compaction PRIORITISE preserving all informatio
    # Main compression entry point
    # ------------------------------------------------------------------

-    def compress(self, messages: List[Dict[str, Any]], current_tokens: int = None, focus_topic: str = None) -> List[Dict[str, Any]]:
+    def compress(self, messages: List[Dict[str, Any]], current_tokens: int = None) -> List[Dict[str, Any]]:
        """Compress conversation messages by summarizing middle turns.

        Algorithm:
@@ -664,12 +632,6 @@ The user has requested that this compaction PRIORITISE preserving all informatio

        After compression, orphaned tool_call / tool_result pairs are cleaned
        up so the API never receives mismatched IDs.
-
-        Args:
-            focus_topic: Optional focus string for guided compression.  When
-                provided, the summariser will prioritise preserving information
-                related to this topic and be more aggressive about compressing
-                everything else.  Inspired by Claude Code's ``/compact``.
        """
        n_messages = len(messages)
        # Only need head + 3 tail messages minimum (token budget decides the real tail size)
@@ -727,7 +689,7 @@ The user has requested that this compaction PRIORITISE preserving all informatio
            )

        # Phase 3: Generate structured summary
-        summary = self._generate_summary(turns_to_summarize, focus_topic=focus_topic)
+        summary = self._generate_summary(turns_to_summarize)

        # Phase 4: Assemble compressed message list
        compressed = []
@@ -782,12 +744,7 @@ The user has requested that this compaction PRIORITISE preserving all informatio
            msg = messages[i].copy()
            if _merge_summary_into_tail and i == compress_end:
                original = msg.get("content") or ""
-                msg["content"] = (
-                    summary
-                    + "\n\n--- END OF CONTEXT SUMMARY — "
-                    "respond to the message below, not the summary above ---\n\n"
-                    + original
-                )
+                msg["content"] = summary + "\n\n" + original
                _merge_summary_into_tail = False
            compressed.append(msg)

@@ -1128,23 +1128,6 @@ def _seed_from_singletons(provider: str, entries: List[PooledCredential]) -> Tup
    elif provider == "openai-codex":
        state = _load_provider_state(auth_store, "openai-codex")
        tokens = state.get("tokens") if isinstance(state, dict) else None
-        # Fallback: import from Codex CLI (~/.codex/auth.json) if Hermes auth
-        # store has no tokens.  This mirrors resolve_codex_runtime_credentials()
-        # so that load_pool() and list_authenticated_providers() detect tokens
-        # that only exist in the Codex CLI shared file.
-        if not (isinstance(tokens, dict) and tokens.get("access_token")):
-            try:
-                from hermes_cli.auth import _import_codex_cli_tokens, _save_codex_tokens
-                cli_tokens = _import_codex_cli_tokens()
-                if cli_tokens:
-                    logger.info("Importing Codex CLI tokens into Hermes auth store.")
-                    _save_codex_tokens(cli_tokens)
-                    # Re-read state after import
-                    auth_store = _load_auth_store()
-                    state = _load_provider_state(auth_store, "openai-codex")
-                    tokens = state.get("tokens") if isinstance(state, dict) else None
-            except Exception as exc:
-                logger.debug("Codex CLI token import failed: %s", exc)
        if isinstance(tokens, dict) and tokens.get("access_token"):
            active_sources.add("device_code")
            changed |= _upsert_entry(
@@ -4,6 +4,7 @@ Pure display functions and classes with no AIAgent dependency.
 Used by AIAgent._execute_tool_calls for CLI feedback.
 """

+import json
 import logging
 import os
 import sys
@@ -13,8 +14,6 @@ from dataclasses import dataclass, field
 from difflib import unified_diff
 from pathlib import Path

-from utils import safe_json_loads
-
 # ANSI escape codes for coloring tool failure indicators
 _RED = "\033[31m"
 _RESET = "\033[0m"
@@ -22,73 +21,11 @@ _RESET = "\033[0m"
 logger = logging.getLogger(__name__)

 _ANSI_RESET = "\033[0m"
-
-# Diff colors — resolved lazily from the skin engine so they adapt
-# to light/dark themes.  Falls back to sensible defaults on import
-# failure.  We cache after first resolution for performance.
-_diff_colors_cached: dict[str, str] | None = None
-
-
-def _diff_ansi() -> dict[str, str]:
-    """Return ANSI escapes for diff display, resolved from the active skin."""
-    global _diff_colors_cached
-    if _diff_colors_cached is not None:
-        return _diff_colors_cached
-
-    # Defaults that work on dark terminals
-    dim = "\033[38;2;150;150;150m"
-    file_c = "\033[38;2;180;160;255m"
-    hunk = "\033[38;2;120;120;140m"
-    minus = "\033[38;2;255;255;255;48;2;120;20;20m"
-    plus = "\033[38;2;255;255;255;48;2;20;90;20m"
-
-    try:
-        from hermes_cli.skin_engine import get_active_skin
-        skin = get_active_skin()
-
-        def _hex_fg(key: str, fallback_rgb: tuple[int, int, int]) -> str:
-            h = skin.get_color(key, "")
-            if h and len(h) == 7 and h[0] == "#":
-                r, g, b = int(h[1:3], 16), int(h[3:5], 16), int(h[5:7], 16)
-                return f"\033[38;2;{r};{g};{b}m"
-            r, g, b = fallback_rgb
-            return f"\033[38;2;{r};{g};{b}m"
-
-        dim = _hex_fg("banner_dim", (150, 150, 150))
-        file_c = _hex_fg("session_label", (180, 160, 255))
-        hunk = _hex_fg("session_border", (120, 120, 140))
-        # minus/plus use background colors — derive from ui_error/ui_ok
-        err_h = skin.get_color("ui_error", "#ef5350")
-        ok_h = skin.get_color("ui_ok", "#4caf50")
-        if err_h and len(err_h) == 7:
-            er, eg, eb = int(err_h[1:3], 16), int(err_h[3:5], 16), int(err_h[5:7], 16)
-            # Use a dark tinted version as background
-            minus = f"\033[38;2;255;255;255;48;2;{max(er//2,20)};{max(eg//4,10)};{max(eb//4,10)}m"
-        if ok_h and len(ok_h) == 7:
-            or_, og, ob = int(ok_h[1:3], 16), int(ok_h[3:5], 16), int(ok_h[5:7], 16)
-            plus = f"\033[38;2;255;255;255;48;2;{max(or_//4,10)};{max(og//2,20)};{max(ob//4,10)}m"
-    except Exception:
-        pass
-
-    _diff_colors_cached = {
-        "dim": dim, "file": file_c, "hunk": hunk,
-        "minus": minus, "plus": plus,
-    }
-    return _diff_colors_cached
-
-
-def reset_diff_colors() -> None:
-    """Reset cached diff colors (call after /skin switch)."""
-    global _diff_colors_cached
-    _diff_colors_cached = None
-
-
-# Module-level helpers — each call resolves from the active skin lazily.
-def _diff_dim():   return _diff_ansi()["dim"]
-def _diff_file():  return _diff_ansi()["file"]
-def _diff_hunk():  return _diff_ansi()["hunk"]
-def _diff_minus(): return _diff_ansi()["minus"]
-def _diff_plus():  return _diff_ansi()["plus"]
+_ANSI_DIM = "\033[38;2;150;150;150m"
+_ANSI_FILE = "\033[38;2;180;160;255m"
+_ANSI_HUNK = "\033[38;2;120;120;140m"
+_ANSI_MINUS = "\033[38;2;255;255;255;48;2;120;20;20m"
+_ANSI_PLUS = "\033[38;2;255;255;255;48;2;20;90;20m"
 _MAX_INLINE_DIFF_FILES = 6
 _MAX_INLINE_DIFF_LINES = 80

@@ -373,8 +310,9 @@ def _result_succeeded(result: str | None) -> bool:
    """Conservatively detect whether a tool result represents success."""
    if not result:
        return False
-    data = safe_json_loads(result)
-    if data is None:
+    try:
+        data = json.loads(result)
+    except (json.JSONDecodeError, TypeError):
        return False
    if not isinstance(data, dict):
        return False
@@ -423,7 +361,10 @@ def extract_edit_diff(
 ) -> str | None:
    """Extract a unified diff from a file-edit tool result."""
    if tool_name == "patch" and result:
-        data = safe_json_loads(result)
+        try:
+            data = json.loads(result)
+        except (json.JSONDecodeError, TypeError):
+            data = None
        if isinstance(data, dict):
            diff = data.get("diff")
            if isinstance(diff, str) and diff.strip():
@@ -462,19 +403,19 @@ def _render_inline_unified_diff(diff: str) -> list[str]:
        if raw_line.startswith("+++ "):
            to_file = raw_line[4:].strip()
            if from_file or to_file:
-                rendered.append(f"{_diff_file()}{from_file or 'a/?'} → {to_file or 'b/?'}{_ANSI_RESET}")
+                rendered.append(f"{_ANSI_FILE}{from_file or 'a/?'} → {to_file or 'b/?'}{_ANSI_RESET}")
            continue
        if raw_line.startswith("@@"):
-            rendered.append(f"{_diff_hunk()}{raw_line}{_ANSI_RESET}")
+            rendered.append(f"{_ANSI_HUNK}{raw_line}{_ANSI_RESET}")
            continue
        if raw_line.startswith("-"):
-            rendered.append(f"{_diff_minus()}{raw_line}{_ANSI_RESET}")
+            rendered.append(f"{_ANSI_MINUS}{raw_line}{_ANSI_RESET}")
            continue
        if raw_line.startswith("+"):
-            rendered.append(f"{_diff_plus()}{raw_line}{_ANSI_RESET}")
+            rendered.append(f"{_ANSI_PLUS}{raw_line}{_ANSI_RESET}")
            continue
        if raw_line.startswith(" "):
-            rendered.append(f"{_diff_dim()}{raw_line}{_ANSI_RESET}")
+            rendered.append(f"{_ANSI_DIM}{raw_line}{_ANSI_RESET}")
            continue
        if raw_line:
            rendered.append(raw_line)
@@ -540,7 +481,7 @@ def _summarize_rendered_diff_sections(
        summary = f"… omitted {omitted_lines} diff line(s)"
        if omitted_files:
            summary += f" across {omitted_files} additional file(s)/section(s)"
-        rendered.append(f"{_diff_hunk()}{summary}{_ANSI_RESET}")
+        rendered.append(f"{_ANSI_HUNK}{summary}{_ANSI_RESET}")

    return rendered

@@ -777,19 +718,23 @@ def _detect_tool_failure(tool_name: str, result: str | None) -> tuple[bool, str]
        return False, ""

    if tool_name == "terminal":
-        data = safe_json_loads(result)
-        if isinstance(data, dict):
+        try:
+            data = json.loads(result)
            exit_code = data.get("exit_code")
            if exit_code is not None and exit_code != 0:
                return True, f" [exit {exit_code}]"
+        except (json.JSONDecodeError, TypeError, AttributeError):
+            logger.debug("Could not parse terminal result as JSON for exit code check")
        return False, ""

    # Memory-specific: distinguish "full" from real errors
    if tool_name == "memory":
-        data = safe_json_loads(result)
-        if isinstance(data, dict):
+        try:
+            data = json.loads(result)
            if data.get("success") is False and "exceed the limit" in data.get("error", ""):
                return True, " [full]"
+        except (json.JSONDecodeError, TypeError, AttributeError):
+            logger.debug("Could not parse memory result as JSON for capacity check")

    # Generic heuristic for non-terminal tools
    lower = result[:500].lower()
@@ -27,14 +27,12 @@ _PROVIDER_PREFIXES: frozenset[str] = frozenset({
    "gemini", "zai", "kimi-coding", "minimax", "minimax-cn", "anthropic", "deepseek",
    "opencode-zen", "opencode-go", "ai-gateway", "kilocode", "alibaba",
    "qwen-oauth",
-    "xiaomi",
    "custom", "local",
    # Common aliases
    "google", "google-gemini", "google-ai-studio",
    "glm", "z-ai", "z.ai", "zhipu", "github", "github-copilot",
    "github-models", "kimi", "moonshot", "claude", "deep-seek",
    "opencode", "zen", "go", "vercel", "kilo", "dashscope", "aliyun", "qwen",
-    "mimo", "xiaomi-mimo",
    "qwen-portal",
 })

@@ -85,11 +83,6 @@ CONTEXT_PROBE_TIERS = [
 # Default context length when no detection method succeeds.
 DEFAULT_FALLBACK_CONTEXT = CONTEXT_PROBE_TIERS[0]

-# Minimum context length required to run Hermes Agent.  Models with fewer
-# tokens cannot maintain enough working memory for tool-calling workflows.
-# Sessions, model switches, and cron jobs should reject models below this.
-MINIMUM_CONTEXT_LENGTH = 64_000
-
 # Thin fallback defaults — only broad model family patterns.
 # These fire only when provider is unknown AND models.dev/OpenRouter/Anthropic
 # all miss. Replaced the previous 80+ entry dict.
@@ -120,14 +113,17 @@ DEFAULT_CONTEXT_LENGTHS = {
    "deepseek": 128000,
    # Meta
    "llama": 131072,
-    # Qwen — specific model families before the catch-all.
-    # Official docs: https://help.aliyun.com/zh/model-studio/developer-reference/
-    "qwen3-coder-plus": 1000000,  # 1M context
-    "qwen3-coder": 262144,        # 256K context
+    # Qwen
    "qwen": 131072,
-    # MiniMax — official docs: 204,800 context for all models
-    # https://platform.minimax.io/docs/api-reference/text-anthropic-api
-    "minimax": 204800,
+    # MiniMax (lowercase — lookup lowercases model names at line 973)
+    "minimax-m1-256k": 1000000,
+    "minimax-m1-128k": 1000000,
+    "minimax-m1-80k": 1000000,
+    "minimax-m1-40k": 1000000,
+    "minimax-m1": 1000000,
+    "minimax-m2.5": 1048576,
+    "minimax-m2.7": 1048576,
+    "minimax": 1048576,
    # GLM
    "glm": 202752,
    # xAI Grok — xAI /v1/models does not return context_length metadata,
@@ -155,11 +151,10 @@ DEFAULT_CONTEXT_LENGTHS = {
    "deepseek-ai/DeepSeek-V3.2": 65536,
    "moonshotai/Kimi-K2.5": 262144,
    "moonshotai/Kimi-K2-Thinking": 262144,
-    "MiniMaxAI/MiniMax-M2.5": 204800,
-    "XiaomiMiMo/MiMo-V2-Flash": 256000,
-    "mimo-v2-pro": 1000000,
-    "mimo-v2-omni": 256000,
-    "mimo-v2-flash": 256000,
+    "MiniMaxAI/MiniMax-M2.5": 1048576,
+    "XiaomiMiMo/MiMo-V2-Flash": 32768,
+    "mimo-v2-pro": 1048576,
+    "mimo-v2-omni": 1048576,
    "zai-org/GLM-5": 202752,
 }

@@ -184,12 +179,6 @@ _MAX_COMPLETION_KEYS = (

 # Local server hostnames / address patterns
 _LOCAL_HOSTS = ("localhost", "127.0.0.1", "::1", "0.0.0.0")
-# Docker / Podman / Lima DNS names that resolve to the host machine
-_CONTAINER_LOCAL_SUFFIXES = (
-    ".docker.internal",
-    ".containers.internal",
-    ".lima.internal",
-)


 def _normalize_base_url(base_url: str) -> str:
@@ -225,8 +214,6 @@ _URL_TO_PROVIDER: Dict[str, str] = {
    "api.fireworks.ai": "fireworks",
    "opencode.ai": "opencode-go",
    "api.x.ai": "xai",
-    "api.xiaomimimo.com": "xiaomi",
-    "xiaomimimo.com": "xiaomi",
 }


@@ -265,9 +252,6 @@ def is_local_endpoint(base_url: str) -> bool:
        return False
    if host in _LOCAL_HOSTS:
        return True
-    # Docker / Podman / Lima internal DNS names (e.g. host.docker.internal)
-    if any(host.endswith(suffix) for suffix in _CONTAINER_LOCAL_SUFFIXES):
-        return True
    # RFC-1918 private ranges and link-local
    import ipaddress
    try:
@@ -1045,21 +1029,16 @@ def get_model_context_length(


 def estimate_tokens_rough(text: str) -> int:
-    """Rough token estimate (~4 chars/token) for pre-flight checks.
-
-    Uses ceiling division so short texts (1-3 chars) never estimate as
-    0 tokens, which would cause the compressor and pre-flight checks to
-    systematically undercount when many short tool results are present.
-    """
+    """Rough token estimate (~4 chars/token) for pre-flight checks."""
    if not text:
        return 0
-    return (len(text) + 3) // 4
+    return len(text) // 4


 def estimate_messages_tokens_rough(messages: List[Dict[str, Any]]) -> int:
    """Rough token estimate for a message list (pre-flight only)."""
    total_chars = sum(len(str(msg)) for msg in messages)
-    return (total_chars + 3) // 4
+    return total_chars // 4


 def estimate_request_tokens_rough(
@@ -1082,4 +1061,4 @@ def estimate_request_tokens_rough(
        total_chars += sum(len(str(msg)) for msg in messages)
    if tools:
        total_chars += len(str(tools))
-    return (total_chars + 3) // 4
+    return total_chars // 4
@@ -161,7 +161,6 @@ PROVIDER_TO_MODELS_DEV: Dict[str, str] = {
    "gemini": "google",
    "google": "google",
    "xai": "xai",
-    "xiaomi": "xiaomi",
    "nvidia": "nvidia",
    "groq": "groq",
    "mistral": "mistral",
@@ -384,14 +383,7 @@ def get_model_capabilities(provider: str, model: str) -> Optional[ModelCapabilit

    # Extract capability flags (default to False if missing)
    supports_tools = bool(entry.get("tool_call", False))
-    # Vision: check both the `attachment` flag and `modalities.input` for "image".
-    # Some models (e.g. gemma-4) list image in input modalities but not attachment.
-    input_mods = entry.get("modalities", {})
-    if isinstance(input_mods, dict):
-        input_mods = input_mods.get("input", [])
-    else:
-        input_mods = []
-    supports_vision = bool(entry.get("attachment", False)) or "image" in input_mods
+    supports_vision = bool(entry.get("attachment", False))
    supports_reasoning = bool(entry.get("reasoning", False))

    # Extract limits
@@ -12,7 +12,7 @@ import threading
 from collections import OrderedDict
 from pathlib import Path

-from hermes_constants import get_hermes_home, get_skills_dir
+from hermes_constants import get_hermes_home
 from typing import Optional

 from agent.skill_utils import (
@@ -548,7 +548,8 @@ def build_skills_system_prompt(
    are read-only — they appear in the index but new skills are always created
    in the local dir.  Local skills take precedence when names collide.
    """
-    skills_dir = get_skills_dir()
+    hermes_home = get_hermes_home()
+    skills_dir = hermes_home / "skills"
    external_dirs = get_all_skills_dirs()[1:]  # skip local (index 0)

    if not skills_dir.exists() and not external_dirs:
@@ -168,7 +168,7 @@ def _build_skill_message(
            subdir_path = skill_dir / subdir
            if subdir_path.exists():
                for f in sorted(subdir_path.rglob("*")):
-                    if f.is_file() and not f.is_symlink():
+                    if f.is_file():
                        rel = str(f.relative_to(skill_dir))
                        supporting.append(rel)

@@ -12,7 +12,7 @@ import sys
 from pathlib import Path
 from typing import Any, Dict, List, Set, Tuple

-from hermes_constants import get_config_path, get_skills_dir
+from hermes_constants import get_hermes_home

 logger = logging.getLogger(__name__)

@@ -130,7 +130,7 @@ def get_disabled_skill_names(platform: str | None = None) -> Set[str]:
    Reads the config file directly (no CLI config imports) to stay
    lightweight.
    """
-    config_path = get_config_path()
+    config_path = get_hermes_home() / "config.yaml"
    if not config_path.exists():
        return set()
    try:
@@ -178,7 +178,7 @@ def get_external_skills_dirs() -> List[Path]:
    path.  Only directories that actually exist are returned.  Duplicates and
    paths that resolve to the local ``~/.hermes/skills/`` are silently skipped.
    """
-    config_path = get_config_path()
+    config_path = get_hermes_home() / "config.yaml"
    if not config_path.exists():
        return []
    try:
@@ -200,7 +200,7 @@ def get_external_skills_dirs() -> List[Path]:
    if not isinstance(raw_dirs, list):
        return []

-    local_skills = get_skills_dir().resolve()
+    local_skills = (get_hermes_home() / "skills").resolve()
    seen: Set[Path] = set()
    result: List[Path] = []

@@ -230,7 +230,7 @@ def get_all_skills_dirs() -> List[Path]:
    The local dir is always first (and always included even if it doesn't exist
    yet — callers handle that).  External dirs follow in config order.
    """
-    dirs = [get_skills_dir()]
+    dirs = [get_hermes_home() / "skills"]
    dirs.extend(get_external_skills_dirs())
    return dirs

@@ -384,7 +384,7 @@ def resolve_skill_config_values(
    current values (or the declared default if the key isn't set).
    Path values are expanded via ``os.path.expanduser``.
    """
-    config_path = get_config_path()
+    config_path = get_hermes_home() / "config.yaml"
    config: Dict[str, Any] = {}
    if config_path.exists():
        try:
@@ -24,7 +24,6 @@ model:
  #   "minimax"      - MiniMax global (requires: MINIMAX_API_KEY)
  #   "minimax-cn"   - MiniMax China (requires: MINIMAX_CN_API_KEY)
  #   "huggingface"  - Hugging Face Inference (requires: HF_TOKEN)
-  #   "xiaomi"       - Xiaomi MiMo (requires: XIAOMI_API_KEY)
  #   "kilocode"     - KiloCode gateway (requires: KILOCODE_API_KEY)
  #   "ai-gateway"   - Vercel AI Gateway (requires: AI_GATEWAY_API_KEY)
  #
@@ -589,7 +588,7 @@ platform_toolsets:
 #   skills_hub   - skill_hub (search/install/manage from online registries — user-driven only)
 #   moa          - mixture_of_agents  (requires OPENROUTER_API_KEY)
 #   todo         - todo (in-memory task planning, no deps)
-#   tts          - text_to_speech  (Edge TTS free, or ELEVENLABS/OPENAI/MINIMAX/MISTRAL key)
+#   tts          - text_to_speech  (Edge TTS free, or ELEVENLABS/OPENAI/MINIMAX key)
 #   cronjob      - cronjob (create/list/update/pause/resume/run/remove scheduled tasks)
 #   rl           - rl_list_environments, rl_start_training, etc. (requires TINKER_API_KEY)
 #
@@ -618,7 +617,7 @@ platform_toolsets:
 #   todo         - Task planning and tracking for multi-step work
 #   memory       - Persistent memory across sessions (personal notes + user profile)
 #   session_search - Search and recall past conversations (FTS5 + Gemini Flash summarization)
-#   tts          - Text-to-speech (Edge TTS free, ElevenLabs, OpenAI, MiniMax, Mistral)
+#   tts          - Text-to-speech (Edge TTS free, ElevenLabs, OpenAI, MiniMax)
 #   cronjob      - Schedule and manage automated tasks (CLI-only)
 #   rl           - RL training tools (Tinker-Atropos)
 #
@@ -774,11 +773,6 @@ display:
  # Toggle at runtime with /verbose in the CLI
  tool_progress: all

-  # Gateway-only natural mid-turn assistant updates.
-  # When true, completed assistant status messages are sent as separate chat
-  # messages. This is independent of tool_progress and gateway streaming.
-  interim_assistant_messages: true
-
  # What Enter does when Hermes is already busy in the CLI.
  #   interrupt: Interrupt the current run and redirect Hermes (default)
  #   queue:     Queue your message for the next turn
@@ -787,7 +781,7 @@ display:

  # Background process notifications (gateway/messaging only).
  # Controls how chatty the process watcher is when you use
-  # terminal(background=true, notify_on_complete=true) from Telegram/Discord/etc.
+  # terminal(background=true, check_interval=...) from Telegram/Discord/etc.
  #   off:     No watcher messages at all
  #   result:  Only the final completion message
  #   error:   Only the final message when exit code != 0
@@ -44,7 +44,7 @@ logger = logging.getLogger(__name__)
 _KNOWN_DELIVERY_PLATFORMS = frozenset({
    "telegram", "discord", "slack", "whatsapp", "signal",
    "matrix", "mattermost", "homeassistant", "dingtalk", "feishu",
-    "wecom", "wecom_callback", "weixin", "sms", "email", "webhook", "bluebubbles",
+    "wecom", "weixin", "sms", "email", "webhook", "bluebubbles",
 })

 from cron.jobs import get_due_jobs, mark_job_run, save_job_output, advance_next_run
@@ -219,21 +219,6 @@ def _deliver_result(job: dict, content: str, adapters=None, loop=None) -> Option
    chat_id = target["chat_id"]
    thread_id = target.get("thread_id")

-    # Diagnostic: log thread_id for topic-aware delivery debugging
-    origin = job.get("origin") or {}
-    origin_thread = origin.get("thread_id")
-    if origin_thread and not thread_id:
-        logger.warning(
-            "Job '%s': origin has thread_id=%s but delivery target lost it "
-            "(deliver=%s, target=%s)",
-            job["id"], origin_thread, job.get("deliver", "local"), target,
-        )
-    elif thread_id:
-        logger.debug(
-            "Job '%s': delivering to %s:%s thread_id=%s",
-            job["id"], platform_name, chat_id, thread_id,
-        )
-
    from tools.send_message_tool import _send_to_platform
    from gateway.config import load_gateway_config, Platform

@@ -249,7 +234,6 @@ def _deliver_result(job: dict, content: str, adapters=None, loop=None) -> Option
        "dingtalk": Platform.DINGTALK,
        "feishu": Platform.FEISHU,
        "wecom": Platform.WECOM,
-        "wecom_callback": Platform.WECOM_CALLBACK,
        "weixin": Platform.WEIXIN,
        "email": Platform.EMAIL,
        "sms": Platform.SMS,
@@ -458,14 +442,6 @@ def _run_job_script(script_path: str) -> tuple[bool, str]:
        stdout = (result.stdout or "").strip()
        stderr = (result.stderr or "").strip()

-        # Redact secrets from both stdout and stderr before any return path.
-        try:
-            from agent.redact import redact_sensitive_text
-            stdout = redact_sensitive_text(stdout)
-            stderr = redact_sensitive_text(stderr)
-        except Exception:
-            pass
-
        if result.returncode != 0:
            parts = [f"Script exited with code {result.returncode}"]
            if stderr:
@@ -474,6 +450,13 @@ def _run_job_script(script_path: str) -> tuple[bool, str]:
                parts.append(f"stdout:\n{stdout}")
            return False, "\n".join(parts)

+        # Redact any secrets that may appear in script output before
+        # they are injected into the LLM prompt context.
+        try:
+            from agent.redact import redact_sensitive_text
+            stdout = redact_sensitive_text(stdout)
+        except Exception:
+            pass
        return True, stdout

    except subprocess.TimeoutExpired:
@@ -641,15 +624,6 @@ def run_job(job: dict) -> tuple[bool, str, str, Optional[str]]:
        except Exception as e:
            logger.warning("Job '%s': failed to load config.yaml, using defaults: %s", job_id, e)

-        # Apply IPv4 preference if configured.
-        try:
-            from hermes_constants import apply_ipv4_preference
-            _net_cfg = _cfg.get("network", {})
-            if isinstance(_net_cfg, dict) and _net_cfg.get("force_ipv4"):
-                apply_ipv4_preference(force=True)
-        except Exception:
-            pass
-
        # Reasoning config from config.yaml
        from hermes_constants import parse_reasoning_effort
        effort = str(_cfg.get("agent", {}).get("reasoning_effort", "")).strip()
@@ -747,7 +721,6 @@ def run_job(job: dict) -> tuple[bool, str, str, Optional[str]]:
            provider_sort=pr.get("sort"),
            disabled_toolsets=["cronjob", "messaging", "clarify"],
            quiet_mode=True,
-            skip_context_files=True,  # Don't inject SOUL.md/AGENTS.md from scheduler cwd
            skip_memory=True,  # Cron system prompts would corrupt user representations
            platform="cron",
            session_id=_cron_session_id,
@@ -5,33 +5,6 @@ set -e
 HERMES_HOME="/opt/data"
 INSTALL_DIR="/opt/hermes"

-# --- Privilege dropping via gosu ---
-# When started as root (the default), optionally remap the hermes user/group
-# to match host-side ownership, fix volume permissions, then re-exec as hermes.
-if [ "$(id -u)" = "0" ]; then
-    if [ -n "$HERMES_UID" ] && [ "$HERMES_UID" != "$(id -u hermes)" ]; then
-        echo "Changing hermes UID to $HERMES_UID"
-        usermod -u "$HERMES_UID" hermes
-    fi
-
-    if [ -n "$HERMES_GID" ] && [ "$HERMES_GID" != "$(id -g hermes)" ]; then
-        echo "Changing hermes GID to $HERMES_GID"
-        groupmod -g "$HERMES_GID" hermes
-    fi
-
-    actual_hermes_uid=$(id -u hermes)
-    if [ "$(stat -c %u "$HERMES_HOME" 2>/dev/null)" != "$actual_hermes_uid" ]; then
-        echo "$HERMES_HOME is not owned by $actual_hermes_uid, fixing"
-        chown -R hermes:hermes "$HERMES_HOME"
-    fi
-
-    echo "Dropping root privileges"
-    exec gosu hermes "$0" "$@"
-fi
-
-# --- Running as hermes from here ---
-source "${INSTALL_DIR}/.venv/bin/activate"
-
 # Create essential directory structure.  Cache and platform directories
 # (cache/images, cache/audio, platforms/whatsapp, etc.) are created on
 # demand by the application — don't pre-create them here so new installs
@@ -11,14 +11,12 @@ When you run `hermes setup` for the first time and Hermes detects `~/.openclaw`,
 ### 2. CLI Command (quick, scriptable)

 ```bash
-hermes claw migrate                      # Preview then migrate (always shows preview first)
-hermes claw migrate --dry-run            # Preview only, no changes
+hermes claw migrate                      # Full migration with confirmation prompt
+hermes claw migrate --dry-run            # Preview what would happen
 hermes claw migrate --preset user-data   # Migrate without API keys/secrets
 hermes claw migrate --yes                # Skip confirmation prompt
 ```

-The migration always shows a full preview of what will be imported before making any changes. You review the preview and confirm before anything is written.
-
 **All options:**

 | Flag | Description |
@@ -41,7 +39,7 @@ Ask the agent to run the migration for you:
 ```

 The agent will use the `openclaw-migration` skill to:
-1. Run a preview first to show what would change
+1. Run a dry-run first to preview changes
 2. Ask about conflict resolution (SOUL.md, skills, etc.)
 3. Let you choose between `user-data` and `full` presets
 4. Execute the migration with your choices
@@ -60,31 +58,16 @@ The agent will use the `openclaw-migration` skill to:
 | Messaging settings | `~/.openclaw/config.yaml` (TELEGRAM_ALLOWED_USERS, MESSAGING_CWD) | `~/.hermes/.env` |
 | TTS assets | `~/.openclaw/workspace/tts/` | `~/.hermes/tts/` |

-Workspace files are also checked at `workspace.default/` and `workspace-main/` as fallback paths (OpenClaw renamed `workspace/` to `workspace-main/` in recent versions).
-
 ### `full` preset (adds to `user-data`)
 | Item | Source | Destination |
 |------|--------|-------------|
-| Telegram bot token | `openclaw.json` channels config | `~/.hermes/.env` |
-| OpenRouter API key | `.env`, `openclaw.json`, or `openclaw.json["env"]` | `~/.hermes/.env` |
-| OpenAI API key | `.env`, `openclaw.json`, or `openclaw.json["env"]` | `~/.hermes/.env` |
-| Anthropic API key | `.env`, `openclaw.json`, or `openclaw.json["env"]` | `~/.hermes/.env` |
-| ElevenLabs API key | `.env`, `openclaw.json`, or `openclaw.json["env"]` | `~/.hermes/.env` |
+| Telegram bot token | `~/.openclaw/config.yaml` | `~/.hermes/.env` |
+| OpenRouter API key | `~/.openclaw/.env` or config | `~/.hermes/.env` |
+| OpenAI API key | `~/.openclaw/.env` or config | `~/.hermes/.env` |
+| Anthropic API key | `~/.openclaw/.env` or config | `~/.hermes/.env` |
+| ElevenLabs API key | `~/.openclaw/.env` or config | `~/.hermes/.env` |

-API keys are searched across four sources: inline config values, `~/.openclaw/.env`, the `openclaw.json` `"env"` sub-object, and per-agent auth profiles.
-
-Only allowlisted secrets are ever imported. Other credentials are skipped and reported.
-
-## OpenClaw Schema Compatibility
-
-The migration handles both old and current OpenClaw config layouts:
-
- **Channel tokens**: Reads from flat paths (`channels.telegram.botToken`) and the newer `accounts.default` layout (`channels.telegram.accounts.default.botToken`)
- **TTS provider**: OpenClaw renamed "edge" to "microsoft" — both are recognized and mapped to Hermes' "edge"
- **Provider API types**: Both short (`openai`, `anthropic`) and hyphenated (`openai-completions`, `anthropic-messages`, `google-generative-ai`) values are mapped correctly
- **thinkingDefault**: All enum values are handled including newer ones (`minimal`, `xhigh`, `adaptive`)
- **Matrix**: Uses `accessToken` field (not `botToken`)
- **SecretRef formats**: Plain strings, env templates (`${VAR}`), and `source: "env"` SecretRefs are resolved. `source: "file"` and `source: "exec"` SecretRefs produce a warning — add those keys manually after migration.
+Only these 6 allowlisted secrets are ever imported. Other credentials are skipped and reported.

 ## Conflict Handling

@@ -101,24 +84,18 @@ For skills, you can also use `--skill-conflict rename` to import conflicting ski

 ## Migration Report

-Every migration produces a report showing:
+Every migration (including dry runs) produces a report showing:
 - **Migrated items** — what was successfully imported
 - **Conflicts** — items skipped because they already exist
 - **Skipped items** — items not found in the source
 - **Errors** — items that failed to import

-For executed migrations, the full report is saved to `~/.hermes/migration/openclaw/<timestamp>/`.
-
-## Post-Migration Notes
-
- **Skills require a new session** — imported skills take effect after restarting your agent or starting a new chat.
- **WhatsApp requires re-pairing** — WhatsApp uses QR-code pairing, not token-based auth. Run `hermes whatsapp` to pair.
- **Archive cleanup** — after migration, you'll be offered to rename `~/.openclaw/` to `.openclaw.pre-migration/` to prevent state confusion. You can also run `hermes claw cleanup` later.
+For execute runs, the full report is saved to `~/.hermes/migration/openclaw/<timestamp>/`.

 ## Troubleshooting

 ### "OpenClaw directory not found"
-The migration looks for `~/.openclaw` by default, then tries `~/.clawdbot` and `~/.moltbot`. If your OpenClaw is installed elsewhere, use `--source`:
+The migration looks for `~/.openclaw` by default. If your OpenClaw is installed elsewhere, use `--source`:
 ```bash
 hermes claw migrate --source /path/to/.openclaw
 ```
@@ -131,12 +108,3 @@ hermes skills install openclaw-migration

 ### Memory overflow
 If your OpenClaw MEMORY.md or USER.md exceeds Hermes' character limits, excess entries are exported to an overflow file in the migration report directory. You can manually review and add the most important ones.
-
-### API keys not found
-Keys might be stored in different places depending on your OpenClaw setup:
- `~/.openclaw/.env` file
- Inline in `openclaw.json` under `models.providers.*.apiKey`
- In `openclaw.json` under the `"env"` or `"env.vars"` sub-objects
- In `~/.openclaw/agents/main/agent/auth-profiles.json`
-
-The migration checks all four. If keys use `source: "file"` or `source: "exec"` SecretRefs, they can't be resolved automatically — add them via `hermes config set`.
@@ -1,329 +0,0 @@
-# Container-Aware CLI Review Fixes Spec
-
-**PR:** NousResearch/hermes-agent#7543
-**Review:** cursor[bot] bugbot review (4094049442) + two prior rounds
-**Date:** 2026-04-12
-**Branch:** `feat/container-aware-cli-clean`
-
-## Review Issues Summary
-
-Six issues were raised across three bugbot review rounds. Three were fixed in intermediate commits (38277a6a, 726cf90f). This spec addresses remaining design concerns surfaced by those reviews and simplifies the implementation based on interview decisions.
-
-| # | Issue | Severity | Status |
-|---|-------|----------|--------|
-| 1 | `os.execvp` retry loop unreachable | Medium | Fixed in 79e8cd12 (switched to subprocess.run) |
-| 2 | Redundant `shutil.which("sudo")` | Medium | Fixed in 38277a6a (reuses `sudo` var) |
-| 3 | Missing `chown -h` on symlink update | Low | Fixed in 38277a6a |
-| 4 | Container routing after `parse_args()` | High | Fixed in 726cf90f |
-| 5 | Hardcoded `/home/${user}` | Medium | Fixed in 726cf90f |
-| 6 | Group membership not gated on `container.enable` | Low | Fixed in 726cf90f |
-
-The mechanical fixes are in place but the overall design needs revision. The retry loop, error swallowing, and process model have deeper issues than what the bugbot flagged.
-
---
-
-## Spec: Revised `_exec_in_container`
-
-### Design Principles
-
-1. **Let it crash.** No silent fallbacks. If `.container-mode` exists but something goes wrong, the error propagates naturally (Python traceback). The only case where container routing is skipped is when `.container-mode` doesn't exist or `HERMES_DEV=1`.
-2. **No retries.** Probe once for sudo, exec once. If it fails, docker/podman's stderr reaches the user verbatim.
-3. **Completely transparent.** No error wrapping, no prefixes, no spinners. Docker's output goes straight through.
-4. **`os.execvp` on the happy path.** Replace the Python process entirely so there's no idle parent during interactive sessions. Note: `execvp` never returns on success (process is replaced) and raises `OSError` on failure (it does not return a value). The container process's exit code becomes the process exit code by definition — no explicit propagation needed.
-5. **One human-readable exception to "let it crash".** `subprocess.TimeoutExpired` from the sudo probe gets a specific catch with a readable message, since a raw traceback for "your Docker daemon is slow" is confusing. All other exceptions propagate naturally.
-
-### Execution Flow
-
-```
-1. get_container_exec_info()
-   - HERMES_DEV=1 → return None (skip routing)
-   - Inside container → return None (skip routing)
-   - .container-mode doesn't exist → return None (skip routing)
-   - .container-mode exists → parse and return dict
-   - .container-mode exists but malformed/unreadable → LET IT CRASH (no try/except)
-
-2. _exec_in_container(container_info, sys.argv[1:])
-   a. shutil.which(backend) → if None, print "{backend} not found on PATH" and sys.exit(1)
-   b. Sudo probe: subprocess.run([runtime, "inspect", "--format", "ok", container_name], timeout=15)
-      - If succeeds → needs_sudo = False
-      - If fails → try subprocess.run([sudo, "-n", runtime, "inspect", ...], timeout=15)
-        - If succeeds → needs_sudo = True
-        - If fails → print error with sudoers hint (including why -n is required) and sys.exit(1)
-      - If TimeoutExpired → catch specifically, print human-readable message about slow daemon
-   c. Build exec_cmd: [sudo? + runtime, "exec", tty_flags, "-u", exec_user, env_flags, container, hermes_bin, *cli_args]
-   d. os.execvp(exec_cmd[0], exec_cmd)
-      - On success: process is replaced — Python is gone, container exit code IS the process exit code
-      - On OSError: let it crash (natural traceback)
-```
-
-### Changes to `hermes_cli/main.py`
-
-#### `_exec_in_container` — rewrite
-
-Remove:
- The entire retry loop (`max_retries`, `for attempt in range(...)`)
- Spinner logic (`"Waiting for container..."`, dots)
- Exit code classification (125/126/127 handling)
- `subprocess.run` for the exec call (keep it only for the sudo probe)
- Special TTY vs non-TTY retry counts
- The `time` import (no longer needed)
-
-Change:
- Use `os.execvp(exec_cmd[0], exec_cmd)` as the final call
- Keep the `subprocess` import only for the sudo probe
- Keep TTY detection for the `-it` vs `-i` flag
- Keep env var forwarding (TERM, COLORTERM, LANG, LC_ALL)
- Keep the sudo probe as-is (it's the one "smart" part)
- Bump probe `timeout` from 5s to 15s — cold podman on a loaded machine needs headroom
- Catch `subprocess.TimeoutExpired` specifically on both probe calls — print a readable message about the daemon being unresponsive instead of a raw traceback
- Expand the sudoers hint error message to explain *why* `-n` (non-interactive) is required: a password prompt would hang the CLI or break piped commands
-
-The function becomes roughly:
-
-```python
-def _exec_in_container(container_info: dict, cli_args: list):
-    """Replace the current process with a command inside the managed container.
-
-    Probes whether sudo is needed (rootful containers), then os.execvp
-    into the container. If exec fails, the OS error propagates naturally.
-    """
-    import shutil
-    import subprocess
-
-    backend = container_info["backend"]
-    container_name = container_info["container_name"]
-    exec_user = container_info["exec_user"]
-    hermes_bin = container_info["hermes_bin"]
-
-    runtime = shutil.which(backend)
-    if not runtime:
-        print(f"Error: {backend} not found on PATH. Cannot route to container.",
-              file=sys.stderr)
-        sys.exit(1)
-
-    # Probe whether we need sudo to see the rootful container.
-    # Timeout is 15s — cold podman on a loaded machine can take a while.
-    # TimeoutExpired is caught specifically for a human-readable message;
-    # all other exceptions propagate naturally.
-    needs_sudo = False
-    sudo = None
-    try:
-        probe = subprocess.run(
-            [runtime, "inspect", "--format", "ok", container_name],
-            capture_output=True, text=True, timeout=15,
-        )
-    except subprocess.TimeoutExpired:
-        print(
-            f"Error: timed out waiting for {backend} to respond.\n"
-            f"The {backend} daemon may be unresponsive or starting up.",
-            file=sys.stderr,
-        )
-        sys.exit(1)
-
-    if probe.returncode != 0:
-        sudo = shutil.which("sudo")
-        if sudo:
-            try:
-                probe2 = subprocess.run(
-                    [sudo, "-n", runtime, "inspect", "--format", "ok", container_name],
-                    capture_output=True, text=True, timeout=15,
-                )
-            except subprocess.TimeoutExpired:
-                print(
-                    f"Error: timed out waiting for sudo {backend} to respond.",
-                    file=sys.stderr,
-                )
-                sys.exit(1)
-
-            if probe2.returncode == 0:
-                needs_sudo = True
-            else:
-                print(
-                    f"Error: container '{container_name}' not found via {backend}.\n"
-                    f"\n"
-                    f"The NixOS service runs the container as root. Your user cannot\n"
-                    f"see it because {backend} uses per-user namespaces.\n"
-                    f"\n"
-                    f"Fix: grant passwordless sudo for {backend}. The -n (non-interactive)\n"
-                    f"flag is required because the CLI calls sudo non-interactively —\n"
-                    f"a password prompt would hang or break piped commands:\n"
-                    f"\n"
-                    f'  security.sudo.extraRules = [{{\n'
-                    f'    users = [ "{os.getenv("USER", "your-user")}" ];\n'
-                    f'    commands = [{{ command = "{runtime}"; options = [ "NOPASSWD" ]; }}];\n'
-                    f'  }}];\n'
-                    f"\n"
-                    f"Or run: sudo hermes {' '.join(cli_args)}",
-                    file=sys.stderr,
-                )
-                sys.exit(1)
-        else:
-            print(
-                f"Error: container '{container_name}' not found via {backend}.\n"
-                f"The container may be running under root. Try: sudo hermes {' '.join(cli_args)}",
-                file=sys.stderr,
-            )
-            sys.exit(1)
-
-    is_tty = sys.stdin.isatty()
-    tty_flags = ["-it"] if is_tty else ["-i"]
-
-    env_flags = []
-    for var in ("TERM", "COLORTERM", "LANG", "LC_ALL"):
-        val = os.environ.get(var)
-        if val:
-            env_flags.extend(["-e", f"{var}={val}"])
-
-    cmd_prefix = [sudo, "-n", runtime] if needs_sudo else [runtime]
-    exec_cmd = (
-        cmd_prefix + ["exec"]
-        + tty_flags
-        + ["-u", exec_user]
-        + env_flags
-        + [container_name, hermes_bin]
-        + cli_args
-    )
-
-    # execvp replaces this process entirely — it never returns on success.
-    # On failure it raises OSError, which propagates naturally.
-    os.execvp(exec_cmd[0], exec_cmd)
-```
-
-#### Container routing call site in `main()` — remove try/except
-
-Current:
-```python
-try:
-    from hermes_cli.config import get_container_exec_info
-    container_info = get_container_exec_info()
-    if container_info:
-        _exec_in_container(container_info, sys.argv[1:])
-        sys.exit(1)  # exec failed if we reach here
-except SystemExit:
-    raise
-except Exception:
-    pass  # Container routing unavailable, proceed locally
-```
-
-Revised:
-```python
-from hermes_cli.config import get_container_exec_info
-container_info = get_container_exec_info()
-if container_info:
-    _exec_in_container(container_info, sys.argv[1:])
-    # Unreachable: os.execvp never returns on success (process is replaced)
-    # and raises OSError on failure (which propagates as a traceback).
-    # This line exists only as a defensive assertion.
-    sys.exit(1)
-```
-
-No try/except. If `.container-mode` doesn't exist, `get_container_exec_info()` returns `None` and we skip routing. If it exists but is broken, the exception propagates with a natural traceback.
-
-Note: `sys.exit(1)` after `_exec_in_container` is dead code in all paths — `os.execvp` either replaces the process or raises. It's kept as a belt-and-suspenders assertion with a comment marking it unreachable, not as actual error handling.
-
-### Changes to `hermes_cli/config.py`
-
-#### `get_container_exec_info` — remove inner try/except
-
-Current code catches `(OSError, IOError)` and returns `None`. This silently hides permission errors, corrupt files, etc.
-
-Change: Remove the try/except around file reading. Keep the early returns for `HERMES_DEV=1` and `_is_inside_container()`. The `FileNotFoundError` from `open()` when `.container-mode` doesn't exist should still return `None` (this is the "container mode not enabled" case). All other exceptions propagate.
-
-```python
-def get_container_exec_info() -> Optional[dict]:
-    if os.environ.get("HERMES_DEV") == "1":
-        return None
-    if _is_inside_container():
-        return None
-
-    container_mode_file = get_hermes_home() / ".container-mode"
-
-    try:
-        with open(container_mode_file, "r") as f:
-            # ... parse key=value lines ...
-    except FileNotFoundError:
-        return None
-    # All other exceptions (PermissionError, malformed data, etc.) propagate
-
-    return { ... }
-```
-
---
-
-## Spec: NixOS Module Changes
-
-### Symlink creation — simplify to two branches
-
-Current: 4 branches (symlink exists, directory exists, other file, doesn't exist).
-
-Revised: 2 branches.
-
-```bash
-if [ -d "${symlinkPath}" ] && [ ! -L "${symlinkPath}" ]; then
-  # Real directory — back it up, then create symlink
-  _backup="${symlinkPath}.bak.$(date +%s)"
-  echo "hermes-agent: backing up existing ${symlinkPath} to $_backup"
-  mv "${symlinkPath}" "$_backup"
-fi
-# For everything else (symlink, doesn't exist, etc.) — just force-create
-ln -sfn "${target}" "${symlinkPath}"
-chown -h ${user}:${cfg.group} "${symlinkPath}"
-```
-
-`ln -sfn` handles: existing symlink (replaces), doesn't exist (creates), and after the `mv` above (creates). The only case that needs special handling is a real directory, because `ln -sfn` cannot atomically replace a directory.
-
-Note: there is a theoretical race between the `[ -d ... ]` check and the `mv` (something could create/remove the directory in between). In practice this is a NixOS activation script running as root during `nixos-rebuild switch` — no other process should be touching `~/.hermes` at that moment. Not worth adding locking for.
-
-### Sudoers — document, don't auto-configure
-
-Do NOT add `security.sudo.extraRules` to the module. Document the sudoers requirement in the module's description/comments and in the error message the CLI prints when sudo probe fails.
-
-### Group membership gating — keep as-is
-
-The fix in 726cf90f (`cfg.container.enable && cfg.container.hostUsers != []`) is correct. Leftover group membership when container mode is disabled is harmless. No cleanup needed.
-
---
-
-## Spec: Test Rewrite
-
-The existing test file (`tests/hermes_cli/test_container_aware_cli.py`) has 16 tests. With the simplified exec model, several are obsolete.
-
-### Tests to keep (update as needed)
-
- `test_is_inside_container_dockerenv` — unchanged
- `test_is_inside_container_containerenv` — unchanged
- `test_is_inside_container_cgroup_docker` — unchanged
- `test_is_inside_container_false_on_host` — unchanged
- `test_get_container_exec_info_returns_metadata` — unchanged
- `test_get_container_exec_info_none_inside_container` — unchanged
- `test_get_container_exec_info_none_without_file` — unchanged
- `test_get_container_exec_info_skipped_when_hermes_dev` — unchanged
- `test_get_container_exec_info_not_skipped_when_hermes_dev_zero` — unchanged
- `test_get_container_exec_info_defaults` — unchanged
- `test_get_container_exec_info_docker_backend` — unchanged
-
-### Tests to add
-
- `test_get_container_exec_info_crashes_on_permission_error` — verify that `PermissionError` propagates (no silent `None` return)
- `test_exec_in_container_calls_execvp` — verify `os.execvp` is called with correct args (runtime, tty flags, user, env, container, binary, cli args)
- `test_exec_in_container_sudo_probe_sets_prefix` — verify that when first probe fails and sudo probe succeeds, `os.execvp` is called with `sudo -n` prefix
- `test_exec_in_container_no_runtime_hard_fails` — keep existing, verify `sys.exit(1)` when `shutil.which` returns None
- `test_exec_in_container_non_tty_uses_i_only` — update to check `os.execvp` args instead of `subprocess.run` args
- `test_exec_in_container_probe_timeout_prints_message` — verify that `subprocess.TimeoutExpired` from the probe produces a human-readable error and `sys.exit(1)`, not a raw traceback
- `test_exec_in_container_container_not_running_no_sudo` — verify the path where runtime exists (`shutil.which` returns a path) but probe returns non-zero and no sudo is available. Should print the "container may be running under root" error. This is distinct from `no_runtime_hard_fails` which covers `shutil.which` returning None.
-
-### Tests to delete
-
- `test_exec_in_container_tty_retries_on_container_failure` — retry loop removed
- `test_exec_in_container_non_tty_retries_silently_exits_126` — retry loop removed
- `test_exec_in_container_propagates_hermes_exit_code` — no subprocess.run to check exit codes; execvp replaces the process. Note: exit code propagation still works correctly — when `os.execvp` succeeds, the container's process *becomes* this process, so its exit code is the process exit code by OS semantics. No application code needed, no test needed. A comment in the function docstring documents this intent for future readers.
-
---
-
-## Out of Scope
-
- Auto-configuring sudoers rules in the NixOS module
- Any changes to `get_container_exec_info` parsing logic beyond the try/except narrowing
- Changes to `.container-mode` file format
- Changes to the `HERMES_DEV=1` bypass
- Changes to container detection logic (`_is_inside_container`)
@@ -49,8 +49,6 @@ class HermesToolCallParser(ToolCallParser):
                    continue

                tc_data = json.loads(raw_json)
-                if "name" not in tc_data:
-                    continue
                tool_calls.append(
                    ChatCompletionMessageToolCall(
                        id=f"call_{uuid.uuid4().hex[:8]}",
@@ -89,8 +89,6 @@ class MistralToolCallParser(ToolCallParser):
                        parsed = [parsed]

                    for tc in parsed:
-                        if "name" not in tc:
-                            continue
                        args = tc.get("arguments", {})
                        if isinstance(args, dict):
                            args = json.dumps(args, ensure_ascii=False)
@@ -63,7 +63,6 @@ class Platform(Enum):
    WEBHOOK = "webhook"
    FEISHU = "feishu"
    WECOM = "wecom"
-    WECOM_CALLBACK = "wecom_callback"
    WEIXIN = "weixin"
    BLUEBUBBLES = "bluebubbles"

@@ -191,7 +190,7 @@ class StreamingConfig:
    """Configuration for real-time token streaming to messaging platforms."""
    enabled: bool = False
    transport: str = "edit"       # "edit" (progressive editMessageText) or "off"
-    edit_interval: float = 1.0    # Seconds between message edits (Telegram rate-limits at ~1/s)
+    edit_interval: float = 0.3    # Seconds between message edits
    buffer_threshold: int = 40    # Chars before forcing an edit
    cursor: str = " ▉"           # Cursor shown during streaming

@@ -211,7 +210,7 @@ class StreamingConfig:
        return cls(
            enabled=data.get("enabled", False),
            transport=data.get("transport", "edit"),
-            edit_interval=float(data.get("edit_interval", 1.0)),
+            edit_interval=float(data.get("edit_interval", 0.3)),
            buffer_threshold=int(data.get("buffer_threshold", 40)),
            cursor=data.get("cursor", " ▉"),
        )
@@ -292,14 +291,9 @@ class GatewayConfig:
            # Feishu uses extra dict for app credentials
            elif platform == Platform.FEISHU and config.extra.get("app_id"):
                connected.append(platform)
-            # WeCom bot mode uses extra dict for bot credentials
+            # WeCom uses extra dict for bot credentials
            elif platform == Platform.WECOM and config.extra.get("bot_id"):
                connected.append(platform)
-            # WeCom callback mode uses corp_id or apps list
-            elif platform == Platform.WECOM_CALLBACK and (
-                config.extra.get("corp_id") or config.extra.get("apps")
-            ):
-                connected.append(platform)
            # BlueBubbles uses extra dict for local server config
            elif platform == Platform.BLUEBUBBLES and config.extra.get("server_url") and config.extra.get("password"):
                connected.append(platform)
@@ -993,23 +987,6 @@ def _apply_env_overrides(config: GatewayConfig) -> None:
                name=os.getenv("WECOM_HOME_CHANNEL_NAME", "Home"),
            )

-    # WeCom callback mode (self-built apps)
-    wecom_callback_corp_id = os.getenv("WECOM_CALLBACK_CORP_ID")
-    wecom_callback_corp_secret = os.getenv("WECOM_CALLBACK_CORP_SECRET")
-    if wecom_callback_corp_id and wecom_callback_corp_secret:
-        if Platform.WECOM_CALLBACK not in config.platforms:
-            config.platforms[Platform.WECOM_CALLBACK] = PlatformConfig()
-        config.platforms[Platform.WECOM_CALLBACK].enabled = True
-        config.platforms[Platform.WECOM_CALLBACK].extra.update({
-            "corp_id": wecom_callback_corp_id,
-            "corp_secret": wecom_callback_corp_secret,
-            "agent_id": os.getenv("WECOM_CALLBACK_AGENT_ID", ""),
-            "token": os.getenv("WECOM_CALLBACK_TOKEN", ""),
-            "encoding_aes_key": os.getenv("WECOM_CALLBACK_ENCODING_AES_KEY", ""),
-            "host": os.getenv("WECOM_CALLBACK_HOST", "0.0.0.0"),
-            "port": int(os.getenv("WECOM_CALLBACK_PORT", "8645")),
-        })
-
    # Weixin (personal WeChat via iLink Bot API)
    weixin_token = os.getenv("WEIXIN_TOKEN")
    weixin_account_id = os.getenv("WEIXIN_ACCOUNT_ID")
@@ -1040,9 +1017,6 @@ def _apply_env_overrides(config: GatewayConfig) -> None:
        weixin_group_allowed_users = os.getenv("WEIXIN_GROUP_ALLOWED_USERS", "").strip()
        if weixin_group_allowed_users:
            extra["group_allow_from"] = weixin_group_allowed_users
-        weixin_split_multiline = os.getenv("WEIXIN_SPLIT_MULTILINE_MESSAGES", "").strip()
-        if weixin_split_multiline:
-            extra["split_multiline_messages"] = weixin_split_multiline
        weixin_home = os.getenv("WEIXIN_HOME_CHANNEL", "").strip()
        if weixin_home:
            config.platforms[Platform.WEIXIN].home_channel = HomeChannel(
@@ -1,206 +0,0 @@
-"""Per-platform display/verbosity configuration resolver.
-
-Provides ``resolve_display_setting()`` — the single entry-point for reading
-display settings with platform-specific overrides and sensible defaults.
-
-Resolution order (first non-None wins):
-    1. ``display.platforms.<platform>.<key>``  — explicit per-platform user override
-    2. ``display.<key>``                       — global user setting
-    3. ``_PLATFORM_DEFAULTS[<platform>][<key>]``  — built-in sensible default
-    4. ``_GLOBAL_DEFAULTS[<key>]``              — built-in global default
-
-Backward compatibility: ``display.tool_progress_overrides`` is still read as a
-fallback for ``tool_progress`` when no ``display.platforms`` entry exists.  A
-config migration (version bump) automatically moves the old format into the new
-``display.platforms`` structure.
-"""
-
-from __future__ import annotations
-
-from typing import Any
-
-# ---------------------------------------------------------------------------
-# Overrideable display settings and their global defaults
-# ---------------------------------------------------------------------------
-# These are the settings that can be configured per-platform.
-# Other display settings (compact, personality, skin, etc.) are CLI-only
-# and don't participate in per-platform resolution.
-
-_GLOBAL_DEFAULTS: dict[str, Any] = {
-    "tool_progress": "all",
-    "show_reasoning": False,
-    "tool_preview_length": 0,
-    "streaming": None,  # None = follow top-level streaming config
-}
-
-# ---------------------------------------------------------------------------
-# Sensible per-platform defaults — tiered by platform capability
-# ---------------------------------------------------------------------------
-# Tier 1 (high): Supports message editing, typically personal/team use
-# Tier 2 (medium): Supports editing but often workspace/customer-facing
-# Tier 3 (low): No edit support — each progress msg is permanent
-# Tier 4 (minimal): Batch/non-interactive delivery
-
-_TIER_HIGH = {
-    "tool_progress": "all",
-    "show_reasoning": False,
-    "tool_preview_length": 40,
-    "streaming": None,  # follow global
-}
-
-_TIER_MEDIUM = {
-    "tool_progress": "new",
-    "show_reasoning": False,
-    "tool_preview_length": 40,
-    "streaming": None,
-}
-
-_TIER_LOW = {
-    "tool_progress": "off",
-    "show_reasoning": False,
-    "tool_preview_length": 40,
-    "streaming": False,
-}
-
-_TIER_MINIMAL = {
-    "tool_progress": "off",
-    "show_reasoning": False,
-    "tool_preview_length": 0,
-    "streaming": False,
-}
-
-_PLATFORM_DEFAULTS: dict[str, dict[str, Any]] = {
-    # Tier 1 — full edit support, personal/team use
-    "telegram":    _TIER_HIGH,
-    "discord":     _TIER_HIGH,
-
-    # Tier 2 — edit support, often customer/workspace channels
-    "slack":           _TIER_MEDIUM,
-    "mattermost":      _TIER_MEDIUM,
-    "matrix":          _TIER_MEDIUM,
-    "feishu":          _TIER_MEDIUM,
-
-    # Tier 3 — no edit support, progress messages are permanent
-    "signal":          _TIER_LOW,
-    "whatsapp":        _TIER_LOW,
-    "bluebubbles":     _TIER_LOW,
-    "weixin":          _TIER_LOW,
-    "wecom":           _TIER_LOW,
-    "wecom_callback":  _TIER_LOW,
-    "dingtalk":        _TIER_LOW,
-
-    # Tier 4 — batch or non-interactive delivery
-    "email":           _TIER_MINIMAL,
-    "sms":             _TIER_MINIMAL,
-    "webhook":         _TIER_MINIMAL,
-    "homeassistant":   _TIER_MINIMAL,
-    "api_server":      {**_TIER_HIGH, "tool_preview_length": 0},
-}
-
-# Canonical set of per-platform overrideable keys (for validation).
-OVERRIDEABLE_KEYS = frozenset(_GLOBAL_DEFAULTS.keys())
-
-
-def resolve_display_setting(
-    user_config: dict,
-    platform_key: str,
-    setting: str,
-    fallback: Any = None,
-) -> Any:
-    """Resolve a display setting with per-platform override support.
-
-    Parameters
-    ----------
-    user_config : dict
-        The full parsed config.yaml dict.
-    platform_key : str
-        Platform config key (e.g. ``"telegram"``, ``"slack"``).  Use
-        ``_platform_config_key(source.platform)`` from gateway/run.py.
-    setting : str
-        Display setting name (e.g. ``"tool_progress"``, ``"show_reasoning"``).
-    fallback : Any
-        Fallback value when the setting isn't found anywhere.
-
-    Returns
-    -------
-    The resolved value, or *fallback* if nothing is configured.
-    """
-    display_cfg = user_config.get("display") or {}
-
-    # 1. Explicit per-platform override (display.platforms.<platform>.<key>)
-    platforms = display_cfg.get("platforms") or {}
-    plat_overrides = platforms.get(platform_key)
-    if isinstance(plat_overrides, dict):
-        val = plat_overrides.get(setting)
-        if val is not None:
-            return _normalise(setting, val)
-
-    # 1b. Backward compat: display.tool_progress_overrides.<platform>
-    if setting == "tool_progress":
-        legacy = display_cfg.get("tool_progress_overrides")
-        if isinstance(legacy, dict):
-            val = legacy.get(platform_key)
-            if val is not None:
-                return _normalise(setting, val)
-
-    # 2. Global user setting (display.<key>)
-    val = display_cfg.get(setting)
-    if val is not None:
-        return _normalise(setting, val)
-
-    # 3. Built-in platform default
-    plat_defaults = _PLATFORM_DEFAULTS.get(platform_key)
-    if plat_defaults:
-        val = plat_defaults.get(setting)
-        if val is not None:
-            return val
-
-    # 4. Built-in global default
-    val = _GLOBAL_DEFAULTS.get(setting)
-    if val is not None:
-        return val
-
-    return fallback
-
-
-def get_platform_defaults(platform_key: str) -> dict[str, Any]:
-    """Return the built-in default display settings for a platform.
-
-    Falls back to ``_GLOBAL_DEFAULTS`` for unknown platforms.
-    """
-    return dict(_PLATFORM_DEFAULTS.get(platform_key, _GLOBAL_DEFAULTS))
-
-
-def get_effective_display(user_config: dict, platform_key: str) -> dict[str, Any]:
-    """Return the fully-resolved display settings for a platform.
-
-    Useful for status commands that want to show all effective settings.
-    """
-    return {
-        key: resolve_display_setting(user_config, platform_key, key)
-        for key in OVERRIDEABLE_KEYS
-    }
-
-
-# ---------------------------------------------------------------------------
-# Helpers
-# ---------------------------------------------------------------------------
-
-def _normalise(setting: str, value: Any) -> Any:
-    """Normalise YAML quirks (bare ``off`` → False in YAML 1.1)."""
-    if setting == "tool_progress":
-        if value is False:
-            return "off"
-        if value is True:
-            return "all"
-        return str(value).lower()
-    if setting in ("show_reasoning", "streaming"):
-        if isinstance(value, str):
-            return value.lower() in ("true", "1", "yes", "on")
-        return bool(value)
-    if setting == "tool_preview_length":
-        try:
-            return int(value)
-        except (TypeError, ValueError):
-            return 0
-    return value
@@ -53,7 +53,6 @@ DEFAULT_HOST = "127.0.0.1"
 DEFAULT_PORT = 8642
 MAX_STORED_RESPONSES = 100
 MAX_REQUEST_BYTES = 1_000_000  # 1 MB default limit for POST bodies
-CHAT_COMPLETIONS_SSE_KEEPALIVE_SECONDS = 30.0


 def check_api_server_requirements() -> bool:
@@ -763,11 +762,7 @@ class APIServerAdapter(BasePlatformAdapter):
        """
        import queue as _q

-        sse_headers = {
-            "Content-Type": "text/event-stream",
-            "Cache-Control": "no-cache",
-            "X-Accel-Buffering": "no",
-        }
+        sse_headers = {"Content-Type": "text/event-stream", "Cache-Control": "no-cache"}
        # CORS middleware can't inject headers into StreamResponse after
        # prepare() flushes them, so resolve CORS headers up front.
        origin = request.headers.get("Origin", "")
@@ -780,8 +775,6 @@ class APIServerAdapter(BasePlatformAdapter):
        await response.prepare(request)

        try:
-            last_activity = time.monotonic()
-
            # Role chunk
            role_chunk = {
                "id": completion_id, "object": "chat.completion.chunk",
@@ -789,7 +782,6 @@ class APIServerAdapter(BasePlatformAdapter):
                "choices": [{"index": 0, "delta": {"role": "assistant"}, "finish_reason": None}],
            }
            await response.write(f"data: {json.dumps(role_chunk)}\n\n".encode())
-            last_activity = time.monotonic()

            # Helper — route a queue item to the correct SSE event.
            async def _emit(item):
@@ -813,7 +805,6 @@ class APIServerAdapter(BasePlatformAdapter):
                        "choices": [{"index": 0, "delta": {"content": item}, "finish_reason": None}],
                    }
                    await response.write(f"data: {json.dumps(content_chunk)}\n\n".encode())
-                return time.monotonic()

            # Stream content chunks as they arrive from the agent
            loop = asyncio.get_event_loop()
@@ -828,19 +819,16 @@ class APIServerAdapter(BasePlatformAdapter):
                                delta = stream_q.get_nowait()
                                if delta is None:
                                    break
-                                last_activity = await _emit(delta)
+                                await _emit(delta)
                            except _q.Empty:
                                break
                        break
-                    if time.monotonic() - last_activity >= CHAT_COMPLETIONS_SSE_KEEPALIVE_SECONDS:
-                        await response.write(b": keepalive\n\n")
-                        last_activity = time.monotonic()
                    continue

                if delta is None:  # End of stream sentinel
                    break

-                last_activity = await _emit(delta)
+                await _emit(delta)

            # Get usage from completed agent
            usage = {"input_tokens": 0, "output_tokens": 0, "total_tokens": 0}
@@ -823,36 +823,7 @@ class BasePlatformAdapter(ABC):
        result = handler(self)
        if asyncio.iscoroutine(result):
            await result
-
-    def _acquire_platform_lock(self, scope: str, identity: str, resource_desc: str) -> bool:
-        """Acquire a scoped lock for this adapter. Returns True on success."""
-        from gateway.status import acquire_scoped_lock
-        self._platform_lock_scope = scope
-        self._platform_lock_identity = identity
-        acquired, existing = acquire_scoped_lock(
-            scope, identity, metadata={'platform': self.platform.value}
-        )
-        if acquired:
-            return True
-        owner_pid = existing.get('pid') if isinstance(existing, dict) else None
-        message = (
-            f'{resource_desc} already in use'
-            + (f' (PID {owner_pid})' if owner_pid else '')
-            + '. Stop the other gateway first.'
-        )
-        logger.error('[%s] %s', self.name, message)
-        self._set_fatal_error(f'{scope}_lock', message, retryable=False)
-        return False
-
-    def _release_platform_lock(self) -> None:
-        """Release the scoped lock acquired by _acquire_platform_lock."""
-        identity = getattr(self, '_platform_lock_identity', None)
-        if not identity:
-            return
-        from gateway.status import release_scoped_lock
-        release_scoped_lock(self._platform_lock_scope, identity)
-        self._platform_lock_identity = None
-
+    
    @property
    def name(self) -> str:
        """Human-readable name for this adapter."""
@@ -30,7 +30,6 @@ from gateway.platforms.base import (
    cache_audio_from_bytes,
    cache_document_from_bytes,
 )
-from gateway.platforms.helpers import strip_markdown

 logger = logging.getLogger(__name__)

@@ -90,7 +89,18 @@ def _normalize_server_url(raw: str) -> str:
    return value.rstrip("/")


-
+def _strip_markdown(text: str) -> str:
+    """Strip common markdown formatting for iMessage plain-text delivery."""
+    text = re.sub(r"\*\*(.+?)\*\*", r"\1", text, flags=re.DOTALL)
+    text = re.sub(r"\*(.+?)\*", r"\1", text, flags=re.DOTALL)
+    text = re.sub(r"__(.+?)__", r"\1", text, flags=re.DOTALL)
+    text = re.sub(r"_(.+?)_", r"\1", text, flags=re.DOTALL)
+    text = re.sub(r"```[a-zA-Z0-9_+-]*\n?", "", text)
+    text = re.sub(r"`(.+?)`", r"\1", text)
+    text = re.sub(r"^#{1,6}\s+", "", text, flags=re.MULTILINE)
+    text = re.sub(r"\[([^\]]+)\]\(([^\)]+)\)", r"\1", text)
+    text = re.sub(r"\n{3,}", "\n\n", text)
+    return text.strip()


 # ---------------------------------------------------------------------------
@@ -383,7 +393,7 @@ class BlueBubblesAdapter(BasePlatformAdapter):
        reply_to: Optional[str] = None,
        metadata: Optional[Dict[str, Any]] = None,
    ) -> SendResult:
-        text = strip_markdown(content or "")
+        text = _strip_markdown(content or "")
        if not text:
            return SendResult(success=False, error="BlueBubbles send requires text")
        chunks = self.truncate_message(text, max_length=self.MAX_MESSAGE_LENGTH)
@@ -669,7 +679,7 @@ class BlueBubblesAdapter(BasePlatformAdapter):
        return info

    def format_message(self, content: str) -> str:
-        return strip_markdown(content)
+        return _strip_markdown(content)

    # ------------------------------------------------------------------
    # Inbound attachment downloading (from #4588)
@@ -42,7 +42,6 @@ except ImportError:
    httpx = None  # type: ignore[assignment]

 from gateway.config import Platform, PlatformConfig
-from gateway.platforms.helpers import MessageDeduplicator
 from gateway.platforms.base import (
    BasePlatformAdapter,
    MessageEvent,
@@ -53,6 +52,8 @@ from gateway.platforms.base import (
 logger = logging.getLogger(__name__)

 MAX_MESSAGE_LENGTH = 20000
+DEDUP_WINDOW_SECONDS = 300
+DEDUP_MAX_SIZE = 1000
 RECONNECT_BACKOFF = [2, 5, 10, 30, 60]
 _SESSION_WEBHOOKS_MAX = 500
 _DINGTALK_WEBHOOK_RE = re.compile(r'^https://api\.dingtalk\.com/')
@@ -88,8 +89,8 @@ class DingTalkAdapter(BasePlatformAdapter):
        self._stream_task: Optional[asyncio.Task] = None
        self._http_client: Optional["httpx.AsyncClient"] = None

-        # Message deduplication
-        self._dedup = MessageDeduplicator(max_size=1000)
+        # Message deduplication: msg_id -> timestamp
+        self._seen_messages: Dict[str, float] = {}
        # Map chat_id -> session_webhook for reply routing
        self._session_webhooks: Dict[str, str] = {}

@@ -169,7 +170,7 @@ class DingTalkAdapter(BasePlatformAdapter):

        self._stream_client = None
        self._session_webhooks.clear()
-        self._dedup.clear()
+        self._seen_messages.clear()
        logger.info("[%s] Disconnected", self.name)

    # -- Inbound message processing -----------------------------------------
@@ -177,7 +178,7 @@ class DingTalkAdapter(BasePlatformAdapter):
    async def _on_message(self, message: "ChatbotMessage") -> None:
        """Process an incoming DingTalk chatbot message."""
        msg_id = getattr(message, "message_id", None) or uuid.uuid4().hex
-        if self._dedup.is_duplicate(msg_id):
+        if self._is_duplicate(msg_id):
            logger.debug("[%s] Duplicate message %s, skipping", self.name, msg_id)
            return

@@ -255,6 +256,20 @@ class DingTalkAdapter(BasePlatformAdapter):
                content = " ".join(parts).strip()
        return content

+    # -- Deduplication ------------------------------------------------------
+
+    def _is_duplicate(self, msg_id: str) -> bool:
+        """Check and record a message ID. Returns True if already seen."""
+        now = time.time()
+        if len(self._seen_messages) > DEDUP_MAX_SIZE:
+            cutoff = now - DEDUP_WINDOW_SECONDS
+            self._seen_messages = {k: v for k, v in self._seen_messages.items() if v > cutoff}
+
+        if msg_id in self._seen_messages:
+            return True
+        self._seen_messages[msg_id] = now
+        return False
+
    # -- Outbound messaging -------------------------------------------------

    async def send(
@@ -45,7 +45,6 @@ sys.path.insert(0, str(_Path(__file__).resolve().parents[2]))
 from gateway.config import Platform, PlatformConfig
 import re

-from gateway.platforms.helpers import MessageDeduplicator, ThreadParticipationTracker
 from gateway.platforms.base import (
    BasePlatformAdapter,
    MessageEvent,
@@ -451,15 +450,18 @@ class DiscordAdapter(BasePlatformAdapter):
        # Track threads where the bot has participated so follow-up messages
        # in those threads don't require @mention.  Persisted to disk so the
        # set survives gateway restarts.
-        self._threads = ThreadParticipationTracker("discord")
+        self._bot_participated_threads: set = self._load_participated_threads()
        # Persistent typing indicator loops per channel (DMs don't reliably
        # show the standard typing gateway event for bots)
        self._typing_tasks: Dict[str, asyncio.Task] = {}
        self._bot_task: Optional[asyncio.Task] = None
-        self._post_connect_task: Optional[asyncio.Task] = None
-        # Dedup cache: prevents duplicate bot responses when Discord
-        # RESUME replays events after reconnects.
-        self._dedup = MessageDeduplicator()
+        # Cap to prevent unbounded growth (Discord threads get archived).
+        self._MAX_TRACKED_THREADS = 500
+        # Dedup cache: message_id → timestamp.  Prevents duplicate bot
+        # responses when Discord RESUME replays events after reconnects.
+        self._seen_messages: Dict[str, float] = {}
+        self._SEEN_TTL = 300   # 5 minutes
+        self._SEEN_MAX = 2000  # prune threshold
        # Reply threading mode: "off" (no replies), "first" (reply on first
        # chunk only, default), "all" (reply-reference on every chunk).
        self._reply_to_mode: str = getattr(config, 'reply_to_mode', 'first') or 'first'
@@ -500,9 +502,18 @@ class DiscordAdapter(BasePlatformAdapter):
            return False

        try:
-            if not self._acquire_platform_lock('discord-bot-token', self.config.token, 'Discord bot token'):
+            # Acquire scoped lock to prevent duplicate bot token usage
+            from gateway.status import acquire_scoped_lock
+            self._token_lock_identity = self.config.token
+            acquired, existing = acquire_scoped_lock('discord-bot-token', self._token_lock_identity, metadata={'platform': 'discord'})
+            if not acquired:
+                owner_pid = existing.get('pid') if isinstance(existing, dict) else None
+                message = f'Discord bot token already in use' + (f' (PID {owner_pid})' if owner_pid else '') + '. Stop the other gateway first.'
+                logger.error('[%s] %s', self.name, message)
+                self._set_fatal_error('discord_token_lock', message, retryable=False)
                return False

+
            # Parse allowed user entries (may contain usernames or IDs)
            allowed_env = os.getenv("DISCORD_ALLOWED_USERS", "")
            if allowed_env:
@@ -546,19 +557,29 @@ class DiscordAdapter(BasePlatformAdapter):

                # Resolve any usernames in the allowed list to numeric IDs
                await adapter_self._resolve_allowed_usernames()
-                adapter_self._ready_event.set()

-                if adapter_self._post_connect_task and not adapter_self._post_connect_task.done():
-                    adapter_self._post_connect_task.cancel()
-                adapter_self._post_connect_task = asyncio.create_task(
-                    adapter_self._run_post_connect_initialization()
-                )
+                # Sync slash commands with Discord
+                try:
+                    synced = await adapter_self._client.tree.sync()
+                    logger.info("[%s] Synced %d slash command(s)", adapter_self.name, len(synced))
+                except Exception as e:  # pragma: no cover - defensive logging
+                    logger.warning("[%s] Slash command sync failed: %s", adapter_self.name, e, exc_info=True)
+                adapter_self._ready_event.set()

            @self._client.event
            async def on_message(message: DiscordMessage):
                # Dedup: Discord RESUME replays events after reconnects (#4777)
-                if adapter_self._dedup.is_duplicate(str(message.id)):
+                msg_id = str(message.id)
+                now = time.time()
+                if msg_id in adapter_self._seen_messages:
                    return
+                adapter_self._seen_messages[msg_id] = now
+                if len(adapter_self._seen_messages) > adapter_self._SEEN_MAX:
+                    cutoff = now - adapter_self._SEEN_TTL
+                    adapter_self._seen_messages = {
+                        k: v for k, v in adapter_self._seen_messages.items()
+                        if v > cutoff
+                    }

                # Always ignore our own messages
                if message.author == self._client.user:
@@ -664,11 +685,23 @@ class DiscordAdapter(BasePlatformAdapter):

        except asyncio.TimeoutError:
            logger.error("[%s] Timeout waiting for connection to Discord", self.name, exc_info=True)
-            self._release_platform_lock()
+            try:
+                from gateway.status import release_scoped_lock
+                if getattr(self, '_token_lock_identity', None):
+                    release_scoped_lock('discord-bot-token', self._token_lock_identity)
+                    self._token_lock_identity = None
+            except Exception:
+                pass
            return False
        except Exception as e:  # pragma: no cover - defensive logging
            logger.error("[%s] Failed to connect to Discord: %s", self.name, e, exc_info=True)
-            self._release_platform_lock()
+            try:
+                from gateway.status import release_scoped_lock
+                if getattr(self, '_token_lock_identity', None):
+                    release_scoped_lock('discord-bot-token', self._token_lock_identity)
+                    self._token_lock_identity = None
+            except Exception:
+                pass
            return False

    async def disconnect(self) -> None:
@@ -686,36 +719,21 @@ class DiscordAdapter(BasePlatformAdapter):
            except Exception as e:  # pragma: no cover - defensive logging
                logger.warning("[%s] Error during disconnect: %s", self.name, e, exc_info=True)

-        if self._post_connect_task and not self._post_connect_task.done():
-            self._post_connect_task.cancel()
-            try:
-                await self._post_connect_task
-            except asyncio.CancelledError:
-                pass
-
        self._running = False
        self._client = None
        self._ready_event.clear()
-        self._post_connect_task = None

-        self._release_platform_lock()
+        # Release the token lock
+        try:
+            from gateway.status import release_scoped_lock
+            if getattr(self, '_token_lock_identity', None):
+                release_scoped_lock('discord-bot-token', self._token_lock_identity)
+                self._token_lock_identity = None
+        except Exception:
+            pass

        logger.info("[%s] Disconnected", self.name)

-    async def _run_post_connect_initialization(self) -> None:
-        """Finish non-critical startup work after Discord is connected."""
-        if not self._client:
-            return
-        try:
-            synced = await asyncio.wait_for(self._client.tree.sync(), timeout=30)
-            logger.info("[%s] Synced %d slash command(s)", self.name, len(synced))
-        except asyncio.TimeoutError:
-            logger.warning("[%s] Slash command sync timed out after 30s", self.name)
-        except asyncio.CancelledError:
-            raise
-        except Exception as e:  # pragma: no cover - defensive logging
-            logger.warning("[%s] Slash command sync failed: %s", self.name, e, exc_info=True)
-
    async def _add_reaction(self, message: Any, emoji: str) -> bool:
        """Add an emoji reaction to a Discord message."""
        if not message or not hasattr(message, "add_reaction"):
@@ -1852,7 +1870,7 @@ class DiscordAdapter(BasePlatformAdapter):

        # Track thread participation so follow-ups don't require @mention
        if thread_id:
-            self._threads.mark(thread_id)
+            self._track_thread(thread_id)

        # If a message was provided, kick off a new Hermes session in the thread
        starter = (message or "").strip()
@@ -2223,6 +2241,49 @@ class DiscordAdapter(BasePlatformAdapter):
            return f"{parent_name} / {thread_name}"
        return thread_name

+    # ------------------------------------------------------------------
+    # Thread participation persistence
+    # ------------------------------------------------------------------
+
+    @staticmethod
+    def _thread_state_path() -> Path:
+        """Path to the persisted thread participation set."""
+        from hermes_cli.config import get_hermes_home
+        return get_hermes_home() / "discord_threads.json"
+
+    @classmethod
+    def _load_participated_threads(cls) -> set:
+        """Load persisted thread IDs from disk."""
+        path = cls._thread_state_path()
+        try:
+            if path.exists():
+                data = json.loads(path.read_text(encoding="utf-8"))
+                if isinstance(data, list):
+                    return set(data)
+        except Exception as e:
+            logger.debug("Could not load discord thread state: %s", e)
+        return set()
+
+    def _save_participated_threads(self) -> None:
+        """Persist the current thread set to disk (best-effort)."""
+        path = self._thread_state_path()
+        try:
+            # Trim to most recent entries if over cap
+            thread_list = list(self._bot_participated_threads)
+            if len(thread_list) > self._MAX_TRACKED_THREADS:
+                thread_list = thread_list[-self._MAX_TRACKED_THREADS:]
+                self._bot_participated_threads = set(thread_list)
+            path.parent.mkdir(parents=True, exist_ok=True)
+            path.write_text(json.dumps(thread_list), encoding="utf-8")
+        except Exception as e:
+            logger.debug("Could not save discord thread state: %s", e)
+
+    def _track_thread(self, thread_id: str) -> None:
+        """Add a thread to the participation set and persist."""
+        if thread_id not in self._bot_participated_threads:
+            self._bot_participated_threads.add(thread_id)
+            self._save_participated_threads()
+
    async def _handle_message(self, message: DiscordMessage) -> None:
        """Handle incoming Discord messages."""
        # In server channels (not DMs), require the bot to be @mentioned
@@ -2274,7 +2335,7 @@ class DiscordAdapter(BasePlatformAdapter):

            # Skip the mention check if the message is in a thread where
            # the bot has previously participated (auto-created or replied in).
-            in_bot_thread = is_thread and thread_id in self._threads
+            in_bot_thread = is_thread and thread_id in self._bot_participated_threads

            if require_mention and not is_free_channel and not in_bot_thread:
                if self._client.user not in message.mentions:
@@ -2300,7 +2361,7 @@ class DiscordAdapter(BasePlatformAdapter):
                    is_thread = True
                    thread_id = str(thread.id)
                    auto_threaded_channel = thread
-                    self._threads.mark(thread_id)
+                    self._track_thread(thread_id)

        # Determine message type
        msg_type = MessageType.TEXT
@@ -2484,7 +2545,7 @@ class DiscordAdapter(BasePlatformAdapter):
        # Track thread participation so the bot won't require @mention for
        # follow-up messages in threads it has already engaged in.
        if thread_id:
-            self._threads.mark(thread_id)
+            self._track_thread(thread_id)

        # Only batch plain text messages — commands, media, etc. dispatch
        # immediately since they won't be split by the Discord client.
@@ -360,21 +360,19 @@ def _render_code_block_element(element: Dict[str, Any]) -> str:


 def _strip_markdown_to_plain_text(text: str) -> str:
-    """Strip markdown formatting to plain text for Feishu text fallbacks.
-
-    Delegates common markdown stripping to the shared helper and adds
-    Feishu-specific patterns (blockquotes, strikethrough, underline tags,
-    horizontal rules, \\r\\n normalisation).
-    """
-    from gateway.platforms.helpers import strip_markdown
    plain = text.replace("\r\n", "\n")
    plain = _MARKDOWN_LINK_RE.sub(lambda m: f"{m.group(1)} ({m.group(2).strip()})", plain)
+    plain = re.sub(r"^#{1,6}\s+", "", plain, flags=re.MULTILINE)
    plain = re.sub(r"^>\s?", "", plain, flags=re.MULTILINE)
    plain = re.sub(r"^\s*---+\s*$", "---", plain, flags=re.MULTILINE)
+    plain = re.sub(r"```(?:[^\n]*\n)?([\s\S]*?)```", lambda m: m.group(1).strip("\n"), plain)
+    plain = re.sub(r"`([^`\n]+)`", r"\1", plain)
+    plain = re.sub(r"\*\*([^*\n]+)\*\*", r"\1", plain)
+    plain = re.sub(r"\*([^*\n]+)\*", r"\1", plain)
    plain = re.sub(r"~~([^~\n]+)~~", r"\1", plain)
    plain = re.sub(r"<u>([\s\S]*?)</u>", r"\1", plain)
-    plain = strip_markdown(plain)
-    return plain
+    plain = re.sub(r"\n{3,}", "\n\n", plain)
+    return plain.strip()


 def _coerce_int(value: Any, default: Optional[int] = None, min_value: int = 0) -> Optional[int]:
@@ -1,261 +0,0 @@
-"""Shared helper classes for gateway platform adapters.
-
-Extracts common patterns that were duplicated across 5-7 adapters:
-message deduplication, text batch aggregation, markdown stripping,
-and thread participation tracking.
-"""
-
-import asyncio
-import json
-import logging
-import re
-import time
-from pathlib import Path
-from typing import TYPE_CHECKING, Dict, Optional
-
-if TYPE_CHECKING:
-    from gateway.platforms.base import BasePlatformAdapter, MessageEvent
-
-logger = logging.getLogger(__name__)
-
-
-# ─── Message Deduplication ────────────────────────────────────────────────────
-
-
-class MessageDeduplicator:
-    """TTL-based message deduplication cache.
-
-    Replaces the identical ``_seen_messages`` / ``_is_duplicate()`` pattern
-    previously duplicated in discord, slack, dingtalk, wecom, weixin,
-    mattermost, and feishu adapters.
-
-    Usage::
-
-        self._dedup = MessageDeduplicator()
-
-        # In message handler:
-        if self._dedup.is_duplicate(msg_id):
-            return
-    """
-
-    def __init__(self, max_size: int = 2000, ttl_seconds: float = 300):
-        self._seen: Dict[str, float] = {}
-        self._max_size = max_size
-        self._ttl = ttl_seconds
-
-    def is_duplicate(self, msg_id: str) -> bool:
-        """Return True if *msg_id* was already seen within the TTL window."""
-        if not msg_id:
-            return False
-        now = time.time()
-        if msg_id in self._seen:
-            return True
-        self._seen[msg_id] = now
-        if len(self._seen) > self._max_size:
-            cutoff = now - self._ttl
-            self._seen = {k: v for k, v in self._seen.items() if v > cutoff}
-        return False
-
-    def clear(self):
-        """Clear all tracked messages."""
-        self._seen.clear()
-
-
-# ─── Text Batch Aggregation ──────────────────────────────────────────────────
-
-
-class TextBatchAggregator:
-    """Aggregates rapid-fire text events into single messages.
-
-    Replaces the ``_enqueue_text_event`` / ``_flush_text_batch`` pattern
-    previously duplicated in telegram, discord, matrix, wecom, and feishu.
-
-    Usage::
-
-        self._text_batcher = TextBatchAggregator(
-            handler=self._message_handler,
-            batch_delay=0.6,
-            split_threshold=1900,
-        )
-
-        # In message dispatch:
-        if msg_type == MessageType.TEXT and self._text_batcher.is_enabled():
-            self._text_batcher.enqueue(event, session_key)
-            return
-    """
-
-    def __init__(
-        self,
-        handler,
-        *,
-        batch_delay: float = 0.6,
-        split_delay: float = 2.0,
-        split_threshold: int = 4000,
-    ):
-        self._handler = handler
-        self._batch_delay = batch_delay
-        self._split_delay = split_delay
-        self._split_threshold = split_threshold
-        self._pending: Dict[str, "MessageEvent"] = {}
-        self._pending_tasks: Dict[str, asyncio.Task] = {}
-
-    def is_enabled(self) -> bool:
-        """Return True if batching is active (delay > 0)."""
-        return self._batch_delay > 0
-
-    def enqueue(self, event: "MessageEvent", key: str) -> None:
-        """Add *event* to the pending batch for *key*."""
-        chunk_len = len(event.text or "")
-        existing = self._pending.get(key)
-        if not existing:
-            event._last_chunk_len = chunk_len  # type: ignore[attr-defined]
-            self._pending[key] = event
-        else:
-            existing.text = f"{existing.text}\n{event.text}"
-            existing._last_chunk_len = chunk_len  # type: ignore[attr-defined]
-
-        # Cancel prior flush timer, start a new one
-        prior = self._pending_tasks.get(key)
-        if prior and not prior.done():
-            prior.cancel()
-        self._pending_tasks[key] = asyncio.create_task(self._flush(key))
-
-    async def _flush(self, key: str) -> None:
-        """Wait then dispatch the batched event for *key*."""
-        current_task = self._pending_tasks.get(key)
-        pending = self._pending.get(key)
-        last_len = getattr(pending, "_last_chunk_len", 0) if pending else 0
-
-        # Use longer delay when the last chunk looks like a split message
-        delay = self._split_delay if last_len >= self._split_threshold else self._batch_delay
-        await asyncio.sleep(delay)
-
-        event = self._pending.pop(key, None)
-        if event:
-            try:
-                await self._handler(event)
-            except Exception:
-                logger.exception("[TextBatchAggregator] Error dispatching batched event for %s", key)
-
-        if self._pending_tasks.get(key) is current_task:
-            self._pending_tasks.pop(key, None)
-
-    def cancel_all(self) -> None:
-        """Cancel all pending flush tasks."""
-        for task in self._pending_tasks.values():
-            if not task.done():
-                task.cancel()
-        self._pending_tasks.clear()
-        self._pending.clear()
-
-
-# ─── Markdown Stripping ──────────────────────────────────────────────────────
-
-# Pre-compiled regexes for performance
-_RE_BOLD = re.compile(r"\*\*(.+?)\*\*", re.DOTALL)
-_RE_ITALIC_STAR = re.compile(r"\*(.+?)\*", re.DOTALL)
-_RE_BOLD_UNDER = re.compile(r"__(.+?)__", re.DOTALL)
-_RE_ITALIC_UNDER = re.compile(r"_(.+?)_", re.DOTALL)
-_RE_CODE_BLOCK = re.compile(r"```[a-zA-Z0-9_+-]*\n?")
-_RE_INLINE_CODE = re.compile(r"`(.+?)`")
-_RE_HEADING = re.compile(r"^#{1,6}\s+", re.MULTILINE)
-_RE_LINK = re.compile(r"\[([^\]]+)\]\([^\)]+\)")
-_RE_MULTI_NEWLINE = re.compile(r"\n{3,}")
-
-
-def strip_markdown(text: str) -> str:
-    """Strip markdown formatting for plain-text platforms (SMS, iMessage, etc.).
-
-    Replaces the identical ``_strip_markdown()`` functions previously
-    duplicated in sms.py, bluebubbles.py, and feishu.py.
-    """
-    text = _RE_BOLD.sub(r"\1", text)
-    text = _RE_ITALIC_STAR.sub(r"\1", text)
-    text = _RE_BOLD_UNDER.sub(r"\1", text)
-    text = _RE_ITALIC_UNDER.sub(r"\1", text)
-    text = _RE_CODE_BLOCK.sub("", text)
-    text = _RE_INLINE_CODE.sub(r"\1", text)
-    text = _RE_HEADING.sub("", text)
-    text = _RE_LINK.sub(r"\1", text)
-    text = _RE_MULTI_NEWLINE.sub("\n\n", text)
-    return text.strip()
-
-
-# ─── Thread Participation Tracking ───────────────────────────────────────────
-
-
-class ThreadParticipationTracker:
-    """Persistent tracking of threads the bot has participated in.
-
-    Replaces the identical ``_load/_save_participated_threads`` +
-    ``_mark_thread_participated`` pattern previously duplicated in
-    discord.py and matrix.py.
-
-    Usage::
-
-        self._threads = ThreadParticipationTracker("discord")
-
-        # Check membership:
-        if thread_id in self._threads:
-            ...
-
-        # Mark participation:
-        self._threads.mark(thread_id)
-    """
-
-    _MAX_TRACKED = 500
-
-    def __init__(self, platform_name: str, max_tracked: int = 500):
-        self._platform = platform_name
-        self._max_tracked = max_tracked
-        self._threads: set = self._load()
-
-    def _state_path(self) -> Path:
-        from hermes_constants import get_hermes_home
-        return get_hermes_home() / f"{self._platform}_threads.json"
-
-    def _load(self) -> set:
-        path = self._state_path()
-        if path.exists():
-            try:
-                return set(json.loads(path.read_text(encoding="utf-8")))
-            except Exception:
-                pass
-        return set()
-
-    def _save(self) -> None:
-        path = self._state_path()
-        path.parent.mkdir(parents=True, exist_ok=True)
-        thread_list = list(self._threads)
-        if len(thread_list) > self._max_tracked:
-            thread_list = thread_list[-self._max_tracked:]
-            self._threads = set(thread_list)
-        path.write_text(json.dumps(thread_list), encoding="utf-8")
-
-    def mark(self, thread_id: str) -> None:
-        """Mark *thread_id* as participated and persist."""
-        if thread_id not in self._threads:
-            self._threads.add(thread_id)
-            self._save()
-
-    def __contains__(self, thread_id: str) -> bool:
-        return thread_id in self._threads
-
-    def clear(self) -> None:
-        self._threads.clear()
-
-
-# ─── Phone Number Redaction ──────────────────────────────────────────────────
-
-
-def redact_phone(phone: str) -> str:
-    """Redact a phone number for logging, preserving country code and last 4.
-
-    Replaces the identical ``_redact_phone()`` functions in signal.py,
-    sms.py, and bluebubbles.py.
-    """
-    if not phone:
-        return "<none>"
-    if len(phone) <= 8:
-        return phone[:2] + "****" + phone[-2:] if len(phone) > 4 else "****"
-    return phone[:4] + "****" + phone[-4:]
@@ -92,7 +92,6 @@ from gateway.platforms.base import (
    ProcessingOutcome,
    SendResult,
 )
-from gateway.platforms.helpers import ThreadParticipationTracker

 logger = logging.getLogger(__name__)

@@ -104,7 +103,7 @@ MAX_MESSAGE_LENGTH = 4000
 # Uses get_hermes_home() so each profile gets its own Matrix store.
 from hermes_constants import get_hermes_dir as _get_hermes_dir
 _STORE_DIR = _get_hermes_dir("platforms/matrix/store", "matrix/store")
-_CRYPTO_DB_PATH = _STORE_DIR / "crypto.db"
+_CRYPTO_PICKLE_PATH = _STORE_DIR / "crypto_store.pickle"

 # Grace period: ignore messages older than this many seconds before startup.
 _STARTUP_GRACE_SECONDS = 5
@@ -165,33 +164,6 @@ def check_matrix_requirements() -> bool:
    return True


-class _CryptoStateStore:
-    """Adapter that satisfies the mautrix crypto StateStore interface.
-
-    OlmMachine requires a StateStore with ``is_encrypted``,
-    ``get_encryption_info``, and ``find_shared_rooms``.  The basic
-    ``MemoryStateStore`` from ``mautrix.client`` doesn't implement these,
-    so we provide simple implementations that consult the client's room
-    state.
-    """
-
-    def __init__(self, client_state_store: Any, joined_rooms: set):
-        self._ss = client_state_store
-        self._joined_rooms = joined_rooms
-
-    async def is_encrypted(self, room_id: str) -> bool:
-        return (await self.get_encryption_info(room_id)) is not None
-
-    async def get_encryption_info(self, room_id: str):
-        if hasattr(self._ss, "get_encryption_info"):
-            return await self._ss.get_encryption_info(room_id)
-        return None
-
-    async def find_shared_rooms(self, user_id: str) -> list:
-        # Return all joined rooms — simple but correct for a single-user bot.
-        return list(self._joined_rooms)
-
-
 class MatrixAdapter(BasePlatformAdapter):
    """Gateway adapter for Matrix (any homeserver)."""

@@ -226,7 +198,6 @@ class MatrixAdapter(BasePlatformAdapter):
        )

        self._client: Any = None  # mautrix.client.Client
-        self._crypto_db: Any = None  # mautrix.util.async_db.Database
        self._sync_task: Optional[asyncio.Task] = None
        self._closing = False
        self._startup_ts: float = 0.0
@@ -245,7 +216,8 @@ class MatrixAdapter(BasePlatformAdapter):
        self._pending_megolm: list = []

        # Thread participation tracking (for require_mention bypass)
-        self._threads = ThreadParticipationTracker("matrix")
+        self._bot_participated_threads: set = self._load_participated_threads()
+        self._MAX_TRACKED_THREADS = 500

        # Mention/thread gating — parsed once from env vars.
        self._require_mention: bool = os.getenv("MATRIX_REQUIRE_MENTION", "true").lower() not in ("false", "0", "no")
@@ -280,92 +252,6 @@ class MatrixAdapter(BasePlatformAdapter):
        self._processed_events_set.add(event_id)
        return False

-    # ------------------------------------------------------------------
-    # E2EE helpers
-    # ------------------------------------------------------------------
-
-    async def _verify_device_keys_on_server(self, client: Any, olm: Any) -> bool:
-        """Verify our device keys are on the homeserver after loading crypto state.
-
-        Returns True if keys are valid or were successfully re-uploaded.
-        Returns False if verification fails (caller should refuse E2EE).
-        """
-        try:
-            resp = await client.query_keys({client.mxid: [client.device_id]})
-        except Exception as exc:
-            logger.error(
-                "Matrix: cannot verify device keys on server: %s — refusing E2EE", exc,
-            )
-            return False
-
-        # query_keys returns typed objects (QueryKeysResponse, DeviceKeys
-        # with KeyID keys).  Normalise to plain strings for comparison.
-        device_keys_map = getattr(resp, "device_keys", {}) or {}
-        our_user_devices = device_keys_map.get(str(client.mxid)) or {}
-        our_keys = our_user_devices.get(str(client.device_id))
-
-        if not our_keys:
-            logger.warning("Matrix: device keys missing from server — re-uploading")
-            olm.account.shared = False
-            try:
-                await olm.share_keys()
-            except Exception as exc:
-                logger.error("Matrix: failed to re-upload device keys: %s", exc)
-                return False
-            return True
-
-        # DeviceKeys.keys is a dict[KeyID, str].  Iterate to find the
-        # ed25519 key rather than constructing a KeyID for lookup.
-        server_ed25519 = None
-        keys_dict = getattr(our_keys, "keys", {}) or {}
-        for key_id, key_value in keys_dict.items():
-            if str(key_id).startswith("ed25519:"):
-                server_ed25519 = str(key_value)
-                break
-        local_ed25519 = olm.account.identity_keys.get("ed25519")
-
-        if server_ed25519 != local_ed25519:
-            if olm.account.shared:
-                # Restored account from DB but server has different keys — corrupted state.
-                logger.error(
-                    "Matrix: server has different identity keys for device %s — "
-                    "local crypto state is stale. Delete %s and restart.",
-                    client.device_id,
-                    _CRYPTO_DB_PATH,
-                )
-                return False
-
-            # Fresh account (never uploaded). Server has stale keys from a
-            # previous installation. Try to delete the old device and re-upload.
-            logger.warning(
-                "Matrix: server has stale keys for device %s — attempting re-upload",
-                client.device_id,
-            )
-            try:
-                await client.api.request(
-                    client.api.Method.DELETE
-                    if hasattr(client.api, "Method")
-                    else "DELETE",
-                    f"/_matrix/client/v3/devices/{client.device_id}",
-                )
-                logger.info("Matrix: deleted stale device %s from server", client.device_id)
-            except Exception:
-                # Device deletion often requires UIA or may simply not be
-                # permitted — that's fine, share_keys will try to overwrite.
-                pass
-            try:
-                await olm.share_keys()
-            except Exception as exc:
-                logger.error(
-                    "Matrix: cannot upload device keys for %s: %s. "
-                    "Try generating a new access token to get a fresh device.",
-                    client.device_id,
-                    exc,
-                )
-                return False
-
-        return True
-
    # ------------------------------------------------------------------
    # Required overrides
    # ------------------------------------------------------------------
@@ -464,54 +350,45 @@ class MatrixAdapter(BasePlatformAdapter):
                return False
            try:
                from mautrix.crypto import OlmMachine
-                from mautrix.crypto.store.asyncpg import PgCryptoStore
-                from mautrix.util.async_db import Database
+                from mautrix.crypto.store import MemoryCryptoStore

-                _STORE_DIR.mkdir(parents=True, exist_ok=True)
+                crypto_store = MemoryCryptoStore()

-                # Remove legacy pickle file from pre-SQLite era.
-                legacy_pickle = _STORE_DIR / "crypto_store.pickle"
-                if legacy_pickle.exists():
-                    logger.info("Matrix: removing legacy crypto_store.pickle (migrated to SQLite)")
-                    legacy_pickle.unlink()
+                # Restore persisted crypto state from a previous run.
+                # Uses HMAC to verify integrity before unpickling.
+                pickle_path = _CRYPTO_PICKLE_PATH
+                if pickle_path.exists():
+                    try:
+                        import hashlib, hmac, pickle
+                        raw = pickle_path.read_bytes()
+                        # Format: 32-byte HMAC-SHA256 signature + pickle data.
+                        if len(raw) > 32:
+                            sig, payload = raw[:32], raw[32:]
+                            # Key is derived from the device_id + user_id (stable per install).
+                            hmac_key = f"{self._user_id}:{self._device_id}".encode()
+                            expected = hmac.new(hmac_key, payload, hashlib.sha256).digest()
+                            if hmac.compare_digest(sig, expected):
+                                saved = pickle.loads(payload)  # noqa: S301
+                                if isinstance(saved, MemoryCryptoStore):
+                                    crypto_store = saved
+                                    logger.info("Matrix: restored E2EE crypto store from %s", pickle_path)
+                            else:
+                                logger.warning("Matrix: crypto store HMAC mismatch — ignoring stale/tampered file")
+                    except Exception as exc:
+                        logger.warning("Matrix: could not restore crypto store: %s", exc)

-                # Open SQLite-backed crypto store.
-                crypto_db = Database.create(
-                    f"sqlite:///{_CRYPTO_DB_PATH}",
-                    upgrade_table=PgCryptoStore.upgrade_table,
-                )
-                await crypto_db.start()
-                self._crypto_db = crypto_db
+                olm = OlmMachine(client, crypto_store, state_store)

-                _acct_id = self._user_id or "hermes"
-                _pickle_key = f"{_acct_id}:{self._device_id or 'default'}"
-                crypto_store = PgCryptoStore(
-                    account_id=_acct_id,
-                    pickle_key=_pickle_key,
-                    db=crypto_db,
-                )
-                await crypto_store.open()
-
-                crypto_state = _CryptoStateStore(state_store, self._joined_rooms)
-                olm = OlmMachine(client, crypto_store, crypto_state)
-
-                # Accept unverified devices so senders share Megolm
-                # session keys with us automatically.
+                # Set trust policy: accept unverified devices so senders
+                # share Megolm session keys with us automatically.
                olm.share_keys_min_trust = TrustState.UNVERIFIED
                olm.send_keys_min_trust = TrustState.UNVERIFIED

                await olm.load()
-
-                # Verify our device keys are still on the homeserver.
-                if not await self._verify_device_keys_on_server(client, olm):
-                    await crypto_db.stop()
-                    await api.session.close()
-                    return False
-
                client.crypto = olm
                logger.info(
                    "Matrix: E2EE enabled (store: %s%s)",
-                    str(_CRYPTO_DB_PATH),
+                    str(_STORE_DIR),
                    f", device_id={client.device_id}" if client.device_id else "",
                )
            except Exception as exc:
@@ -541,26 +418,12 @@ class MatrixAdapter(BasePlatformAdapter):
            if isinstance(sync_data, dict):
                rooms_join = sync_data.get("rooms", {}).get("join", {})
                self._joined_rooms = set(rooms_join.keys())
-                # Store the next_batch token so incremental syncs start
-                # from where the initial sync left off.
-                nb = sync_data.get("next_batch")
-                if nb:
-                    await client.sync_store.put_next_batch(nb)
                logger.info(
                    "Matrix: initial sync complete, joined %d rooms",
                    len(self._joined_rooms),
                )
                # Build DM room cache from m.direct account data.
                await self._refresh_dm_cache()
-
-                # Dispatch events from the initial sync so the OlmMachine
-                # receives to-device key shares queued while we were offline.
-                try:
-                    tasks = client.handle_sync(sync_data)
-                    if tasks:
-                        await asyncio.gather(*tasks)
-                except Exception as exc:
-                    logger.warning("Matrix: initial sync event dispatch error: %s", exc)
            else:
                logger.warning("Matrix: initial sync returned unexpected type %s", type(sync_data).__name__)
        except Exception as exc:
@@ -589,12 +452,21 @@ class MatrixAdapter(BasePlatformAdapter):
            except (asyncio.CancelledError, Exception):
                pass

-        # Close the SQLite crypto store database.
-        if hasattr(self, "_crypto_db") and self._crypto_db:
+        # Persist E2EE crypto store before closing so the next restart
+        # can decrypt events using sessions from this run.
+        if self._client and self._encryption and getattr(self._client, "crypto", None):
            try:
-                await self._crypto_db.stop()
+                import hashlib, hmac, pickle
+                crypto_store = self._client.crypto.crypto_store
+                _STORE_DIR.mkdir(parents=True, exist_ok=True)
+                pickle_path = _CRYPTO_PICKLE_PATH
+                payload = pickle.dumps(crypto_store)
+                hmac_key = f"{self._user_id}:{self._device_id}".encode()
+                sig = hmac.new(hmac_key, payload, hashlib.sha256).digest()
+                pickle_path.write_bytes(sig + payload)
+                logger.info("Matrix: persisted E2EE crypto store to %s", pickle_path)
            except Exception as exc:
-                logger.debug("Matrix: could not close crypto DB on disconnect: %s", exc)
+                logger.debug("Matrix: could not persist crypto store on disconnect: %s", exc)

        if self._client:
            try:
@@ -937,35 +809,21 @@ class MatrixAdapter(BasePlatformAdapter):

    async def _sync_loop(self) -> None:
        """Continuously sync with the homeserver."""
-        client = self._client
-        # Resume from the token stored during the initial sync.
-        next_batch = await client.sync_store.get_next_batch()
        while not self._closing:
            try:
-                sync_data = await client.sync(
-                    since=next_batch, timeout=30000,
-                )
+                sync_data = await self._client.sync(timeout=30000)
                if isinstance(sync_data, dict):
                    # Update joined rooms from sync response.
                    rooms_join = sync_data.get("rooms", {}).get("join", {})
                    if rooms_join:
                        self._joined_rooms.update(rooms_join.keys())

-                    # Advance the sync token so the next request is
-                    # incremental instead of a full initial sync.
-                    nb = sync_data.get("next_batch")
-                    if nb:
-                        next_batch = nb
-                        await client.sync_store.put_next_batch(nb)
-
-                    # Dispatch events to registered handlers so that
-                    # _on_room_message / _on_reaction / _on_invite fire.
+                # Share keys periodically if E2EE is enabled.
+                if self._encryption and getattr(self._client, "crypto", None):
                    try:
-                        tasks = client.handle_sync(sync_data)
-                        if tasks:
-                            await asyncio.gather(*tasks)
+                        await self._client.crypto.share_keys()
                    except Exception as exc:
-                        logger.warning("Matrix: sync event dispatch error: %s", exc)
+                        logger.warning("Matrix: E2EE key share failed: %s", exc)

                # Retry any buffered undecrypted events.
                if self._pending_megolm:
@@ -1126,7 +984,7 @@ class MatrixAdapter(BasePlatformAdapter):
        # Require-mention gating.
        if not is_dm:
            is_free_room = room_id in self._free_rooms
-            in_bot_thread = bool(thread_id and thread_id in self._threads)
+            in_bot_thread = bool(thread_id and thread_id in self._bot_participated_threads)
            if self._require_mention and not is_free_room and not in_bot_thread:
                if not is_mentioned:
                    return None
@@ -1134,7 +992,7 @@ class MatrixAdapter(BasePlatformAdapter):
        # DM mention-thread.
        if is_dm and not thread_id and self._dm_mention_threads and is_mentioned:
            thread_id = event_id
-            self._threads.mark(thread_id)
+            self._track_thread(thread_id)

        # Strip mention from body.
        if is_mentioned:
@@ -1143,7 +1001,7 @@ class MatrixAdapter(BasePlatformAdapter):
        # Auto-thread.
        if not is_dm and not thread_id and self._auto_thread:
            thread_id = event_id
-            self._threads.mark(thread_id)
+            self._track_thread(thread_id)

        display_name = await self._get_display_name(room_id, sender)
        source = self.build_source(
@@ -1155,7 +1013,7 @@ class MatrixAdapter(BasePlatformAdapter):
        )

        if thread_id:
-            self._threads.mark(thread_id)
+            self._track_thread(thread_id)

        self._background_read_receipt(room_id, event_id)

@@ -1804,6 +1662,48 @@ class MatrixAdapter(BasePlatformAdapter):
            for rid in self._joined_rooms
        }

+    # ------------------------------------------------------------------
+    # Thread participation tracking
+    # ------------------------------------------------------------------
+
+    @staticmethod
+    def _thread_state_path() -> Path:
+        """Path to the persisted thread participation set."""
+        from hermes_cli.config import get_hermes_home
+        return get_hermes_home() / "matrix_threads.json"
+
+    @classmethod
+    def _load_participated_threads(cls) -> set:
+        """Load persisted thread IDs from disk."""
+        path = cls._thread_state_path()
+        try:
+            if path.exists():
+                data = json.loads(path.read_text(encoding="utf-8"))
+                if isinstance(data, list):
+                    return set(data)
+        except Exception as e:
+            logger.debug("Could not load matrix thread state: %s", e)
+        return set()
+
+    def _save_participated_threads(self) -> None:
+        """Persist the current thread set to disk (best-effort)."""
+        path = self._thread_state_path()
+        try:
+            thread_list = list(self._bot_participated_threads)
+            if len(thread_list) > self._MAX_TRACKED_THREADS:
+                thread_list = thread_list[-self._MAX_TRACKED_THREADS:]
+                self._bot_participated_threads = set(thread_list)
+            path.parent.mkdir(parents=True, exist_ok=True)
+            path.write_text(json.dumps(thread_list), encoding="utf-8")
+        except Exception as e:
+            logger.debug("Could not save matrix thread state: %s", e)
+
+    def _track_thread(self, thread_id: str) -> None:
+        """Add a thread to the participation set and persist."""
+        if thread_id not in self._bot_participated_threads:
+            self._bot_participated_threads.add(thread_id)
+            self._save_participated_threads()
+
    # ------------------------------------------------------------------
    # Mention detection helpers
    # ------------------------------------------------------------------
@@ -18,11 +18,11 @@ import json
 import logging
 import os
 import re
+import time
 from pathlib import Path
 from typing import Any, Dict, List, Optional

 from gateway.config import Platform, PlatformConfig
-from gateway.platforms.helpers import MessageDeduplicator
 from gateway.platforms.base import (
    BasePlatformAdapter,
    MessageEvent,
@@ -96,8 +96,10 @@ class MattermostAdapter(BasePlatformAdapter):
            or os.getenv("MATTERMOST_REPLY_MODE", "off")
        ).lower()

-        # Dedup cache (prevent reprocessing)
-        self._dedup = MessageDeduplicator()
+        # Dedup cache: post_id → timestamp (prevent reprocessing)
+        self._seen_posts: Dict[str, float] = {}
+        self._SEEN_MAX = 2000
+        self._SEEN_TTL = 300  # 5 minutes

    # ------------------------------------------------------------------
    # HTTP helpers
@@ -602,8 +604,10 @@ class MattermostAdapter(BasePlatformAdapter):
        post_id = post.get("id", "")

        # Dedup.
-        if self._dedup.is_duplicate(post_id):
+        self._prune_seen()
+        if post_id in self._seen_posts:
            return
+        self._seen_posts[post_id] = time.time()

        # Build message event.
        channel_id = post.get("channel_id", "")
@@ -730,4 +734,13 @@ class MattermostAdapter(BasePlatformAdapter):

        await self.handle_message(msg_event)

-
+    def _prune_seen(self) -> None:
+        """Remove expired entries from the dedup cache."""
+        if len(self._seen_posts) < self._SEEN_MAX:
+            return
+        now = time.time()
+        self._seen_posts = {
+            pid: ts
+            for pid, ts in self._seen_posts.items()
+            if now - ts < self._SEEN_TTL
+        }
@@ -37,7 +37,6 @@ from gateway.platforms.base import (
    cache_document_from_bytes,
    cache_image_from_url,
 )
-from gateway.platforms.helpers import redact_phone

 logger = logging.getLogger(__name__)

@@ -52,10 +51,22 @@ SSE_RETRY_DELAY_MAX = 60.0
 HEALTH_CHECK_INTERVAL = 30.0  # seconds between health checks
 HEALTH_CHECK_STALE_THRESHOLD = 120.0  # seconds without SSE activity before concern

+# E.164 phone number pattern for redaction
+_PHONE_RE = re.compile(r"\+[1-9]\d{6,14}")
+
+
 # ---------------------------------------------------------------------------
 # Helpers
 # ---------------------------------------------------------------------------

+def _redact_phone(phone: str) -> str:
+    """Redact a phone number for logging: +15551234567 -> +155****4567."""
+    if not phone:
+        return "<none>"
+    if len(phone) <= 8:
+        return phone[:2] + "****" + phone[-2:] if len(phone) > 4 else "****"
+    return phone[:4] + "****" + phone[-4:]
+

 def _parse_comma_list(value: str) -> List[str]:
    """Split a comma-separated string into a list, stripping whitespace."""
@@ -173,8 +184,10 @@ class SignalAdapter(BasePlatformAdapter):
        self._recent_sent_timestamps: set = set()
        self._max_recent_timestamps = 50

+        self._phone_lock_identity: Optional[str] = None
+
        logger.info("Signal adapter initialized: url=%s account=%s groups=%s",
-                     self.http_url, redact_phone(self.account),
+                     self.http_url, _redact_phone(self.account),
                     "enabled" if self.group_allow_from else "disabled")

    # ------------------------------------------------------------------
@@ -189,7 +202,23 @@ class SignalAdapter(BasePlatformAdapter):

        # Acquire scoped lock to prevent duplicate Signal listeners for the same phone
        try:
-            if not self._acquire_platform_lock('signal-phone', self.account, 'Signal account'):
+            from gateway.status import acquire_scoped_lock
+
+            self._phone_lock_identity = self.account
+            acquired, existing = acquire_scoped_lock(
+                "signal-phone",
+                self._phone_lock_identity,
+                metadata={"platform": self.platform.value},
+            )
+            if not acquired:
+                owner_pid = existing.get("pid") if isinstance(existing, dict) else None
+                message = (
+                    "Another local Hermes gateway is already using this Signal account"
+                    + (f" (PID {owner_pid})." if owner_pid else ".")
+                    + " Stop the other gateway before starting a second Signal listener."
+                )
+                logger.error("Signal: %s", message)
+                self._set_fatal_error("signal_phone_lock", message, retryable=False)
                return False
        except Exception as e:
            logger.warning("Signal: Could not acquire phone lock (non-fatal): %s", e)
@@ -241,7 +270,13 @@ class SignalAdapter(BasePlatformAdapter):
            await self.client.aclose()
            self.client = None

-        self._release_platform_lock()
+        if self._phone_lock_identity:
+            try:
+                from gateway.status import release_scoped_lock
+                release_scoped_lock("signal-phone", self._phone_lock_identity)
+            except Exception as e:
+                logger.warning("Signal: Error releasing phone lock: %s", e, exc_info=True)
+            self._phone_lock_identity = None

        logger.info("Signal: disconnected")

@@ -507,7 +542,7 @@ class SignalAdapter(BasePlatformAdapter):
        )

        logger.debug("Signal: message from %s in %s: %s",
-                      redact_phone(sender), chat_id[:20], (text or "")[:50])
+                      _redact_phone(sender), chat_id[:20], (text or "")[:50])

        await self.handle_message(event)

@@ -33,7 +33,6 @@ from pathlib import Path as _Path
 sys.path.insert(0, str(_Path(__file__).resolve().parents[2]))

 from gateway.config import Platform, PlatformConfig
-from gateway.platforms.helpers import MessageDeduplicator
 from gateway.platforms.base import (
    BasePlatformAdapter,
    MessageEvent,
@@ -90,9 +89,11 @@ class SlackAdapter(BasePlatformAdapter):
        self._team_clients: Dict[str, AsyncWebClient] = {}   # team_id → WebClient
        self._team_bot_user_ids: Dict[str, str] = {}          # team_id → bot_user_id
        self._channel_team: Dict[str, str] = {}                # channel_id → team_id
-        # Dedup cache: prevents duplicate bot responses when Socket Mode
-        # reconnects redeliver events.
-        self._dedup = MessageDeduplicator()
+        # Dedup cache: event_ts → timestamp.  Prevents duplicate bot
+        # responses when Socket Mode reconnects redeliver events.
+        self._seen_messages: Dict[str, float] = {}
+        self._SEEN_TTL = 300   # 5 minutes
+        self._SEEN_MAX = 2000  # prune threshold
        # Track pending approval message_ts → resolved flag to prevent
        # double-clicks on approval buttons.
        self._approval_resolved: Dict[str, bool] = {}
@@ -151,7 +152,15 @@ class SlackAdapter(BasePlatformAdapter):
                logger.warning("[Slack] Failed to read %s: %s", tokens_file, e)

        try:
-            if not self._acquire_platform_lock('slack-app-token', app_token, 'Slack app token'):
+            # Acquire scoped lock to prevent duplicate app token usage
+            from gateway.status import acquire_scoped_lock
+            self._token_lock_identity = app_token
+            acquired, existing = acquire_scoped_lock('slack-app-token', app_token, metadata={'platform': 'slack'})
+            if not acquired:
+                owner_pid = existing.get('pid') if isinstance(existing, dict) else None
+                message = f'Slack app token already in use' + (f' (PID {owner_pid})' if owner_pid else '') + '. Stop the other gateway first.'
+                logger.error('[%s] %s', self.name, message)
+                self._set_fatal_error('slack_token_lock', message, retryable=False)
                return False

            # First token is the primary — used for AsyncApp / Socket Mode
@@ -238,7 +247,14 @@ class SlackAdapter(BasePlatformAdapter):
                logger.warning("[Slack] Error while closing Socket Mode handler: %s", e, exc_info=True)
        self._running = False

-        self._release_platform_lock()
+        # Release the token lock (use stored identity, not re-read env)
+        try:
+            from gateway.status import release_scoped_lock
+            if getattr(self, '_token_lock_identity', None):
+                release_scoped_lock('slack-app-token', self._token_lock_identity)
+                self._token_lock_identity = None
+        except Exception:
+            pass

        logger.info("[Slack] Disconnected")

@@ -937,8 +953,17 @@ class SlackAdapter(BasePlatformAdapter):
        """Handle an incoming Slack message event."""
        # Dedup: Slack Socket Mode can redeliver events after reconnects (#4777)
        event_ts = event.get("ts", "")
-        if event_ts and self._dedup.is_duplicate(event_ts):
-            return
+        if event_ts:
+            now = time.time()
+            if event_ts in self._seen_messages:
+                return
+            self._seen_messages[event_ts] = now
+            if len(self._seen_messages) > self._SEEN_MAX:
+                cutoff = now - self._SEEN_TTL
+                self._seen_messages = {
+                    k: v for k, v in self._seen_messages.items()
+                    if v > cutoff
+                }

        # Bot message filtering (SLACK_ALLOW_BOTS / config allow_bots):
        #   "none"     — ignore all bot messages (default, backward-compatible)
@@ -10,9 +10,6 @@ Shares credentials with the optional telephony skill — same env vars:

 Gateway-specific env vars:
  - SMS_WEBHOOK_PORT     (default 8080)
-  - SMS_WEBHOOK_HOST     (default 0.0.0.0)
-  - SMS_WEBHOOK_URL      (public URL for Twilio signature validation — required)
-  - SMS_INSECURE_NO_SIGNATURE  (true to disable signature validation — dev only)
  - SMS_ALLOWED_USERS    (comma-separated E.164 phone numbers)
  - SMS_ALLOW_ALL_USERS  (true/false)
  - SMS_HOME_CHANNEL     (phone number for cron delivery)
@@ -20,10 +17,9 @@ Gateway-specific env vars:

 import asyncio
 import base64
-import hashlib
-import hmac
 import logging
 import os
+import re
 import urllib.parse
 from typing import Any, Dict, Optional

@@ -34,14 +30,24 @@ from gateway.platforms.base import (
    MessageType,
    SendResult,
 )
-from gateway.platforms.helpers import redact_phone, strip_markdown

 logger = logging.getLogger(__name__)

 TWILIO_API_BASE = "https://api.twilio.com/2010-04-01/Accounts"
 MAX_SMS_LENGTH = 1600  # ~10 SMS segments
 DEFAULT_WEBHOOK_PORT = 8080
-DEFAULT_WEBHOOK_HOST = "0.0.0.0"
+
+# E.164 phone number pattern for redaction
+_PHONE_RE = re.compile(r"\+[1-9]\d{6,14}")
+
+
+def _redact_phone(phone: str) -> str:
+    """Redact a phone number for logging: +15551234567 -> +1555***4567."""
+    if not phone:
+        return "<none>"
+    if len(phone) <= 8:
+        return phone[:2] + "***" + phone[-2:] if len(phone) > 4 else "****"
+    return phone[:5] + "***" + phone[-4:]


 def check_sms_requirements() -> bool:
@@ -71,8 +77,6 @@ class SmsAdapter(BasePlatformAdapter):
        self._webhook_port: int = int(
            os.getenv("SMS_WEBHOOK_PORT", str(DEFAULT_WEBHOOK_PORT))
        )
-        self._webhook_host: str = os.getenv("SMS_WEBHOOK_HOST", DEFAULT_WEBHOOK_HOST)
-        self._webhook_url: str = os.getenv("SMS_WEBHOOK_URL", "").strip()
        self._runner = None
        self._http_session: Optional["aiohttp.ClientSession"] = None

@@ -94,33 +98,13 @@ class SmsAdapter(BasePlatformAdapter):
            logger.error("[sms] TWILIO_PHONE_NUMBER not set — cannot send replies")
            return False

-        insecure_no_sig = os.getenv("SMS_INSECURE_NO_SIGNATURE", "").lower() == "true"
-
-        if not self._webhook_url and not insecure_no_sig:
-            logger.error(
-                "[sms] Refusing to start: SMS_WEBHOOK_URL is required for Twilio "
-                "signature validation. Set it to the public URL configured in your "
-                "Twilio console (e.g. https://example.com/webhooks/twilio). "
-                "For local development without validation, set "
-                "SMS_INSECURE_NO_SIGNATURE=true (NOT recommended for production).",
-            )
-            return False
-
-        if insecure_no_sig and not self._webhook_url:
-            logger.warning(
-                "[sms] SMS_INSECURE_NO_SIGNATURE=true — Twilio signature validation "
-                "is DISABLED. Any client that can reach port %d can inject messages. "
-                "Do NOT use this in production.",
-                self._webhook_port,
-            )
-
        app = web.Application()
        app.router.add_post("/webhooks/twilio", self._handle_webhook)
        app.router.add_get("/health", lambda _: web.Response(text="ok"))

        self._runner = web.AppRunner(app)
        await self._runner.setup()
-        site = web.TCPSite(self._runner, self._webhook_host, self._webhook_port)
+        site = web.TCPSite(self._runner, "0.0.0.0", self._webhook_port)
        await site.start()
        self._http_session = aiohttp.ClientSession(
            timeout=aiohttp.ClientTimeout(total=30),
@@ -128,10 +112,9 @@ class SmsAdapter(BasePlatformAdapter):
        self._running = True

        logger.info(
-            "[sms] Twilio webhook server listening on %s:%d, from: %s",
-            self._webhook_host,
+            "[sms] Twilio webhook server listening on port %d, from: %s",
            self._webhook_port,
-            redact_phone(self._from_number),
+            _redact_phone(self._from_number),
        )
        return True

@@ -180,7 +163,7 @@ class SmsAdapter(BasePlatformAdapter):
                            error_msg = body.get("message", str(body))
                            logger.error(
                                "[sms] send failed to %s: %s %s",
-                                redact_phone(chat_id),
+                                _redact_phone(chat_id),
                                resp.status,
                                error_msg,
                            )
@@ -191,7 +174,7 @@ class SmsAdapter(BasePlatformAdapter):
                        msg_sid = body.get("sid", "")
                        last_result = SendResult(success=True, message_id=msg_sid)
                except Exception as e:
-                    logger.error("[sms] send error to %s: %s", redact_phone(chat_id), e)
+                    logger.error("[sms] send error to %s: %s", _redact_phone(chat_id), e)
                    return SendResult(success=False, error=str(e))
        finally:
            # Close session only if we created a fallback (no persistent session)
@@ -209,75 +192,16 @@ class SmsAdapter(BasePlatformAdapter):

    def format_message(self, content: str) -> str:
        """Strip markdown — SMS renders it as literal characters."""
-        return strip_markdown(content)
-
-    # ------------------------------------------------------------------
-    # Twilio signature validation
-    # ------------------------------------------------------------------
-
-    def _validate_twilio_signature(
-        self, url: str, post_params: dict, signature: str,
-    ) -> bool:
-        """Validate ``X-Twilio-Signature`` header (HMAC-SHA1, base64).
-
-        Tries both with and without the default port for the URL scheme,
-        since Twilio may sign with either variant.
-
-        Algorithm: https://www.twilio.com/docs/usage/security#validating-requests
-        """
-        if self._check_signature(url, post_params, signature):
-            return True
-
-        variant = self._port_variant_url(url)
-        if variant and self._check_signature(variant, post_params, signature):
-            return True
-
-        return False
-
-    def _check_signature(
-        self, url: str, post_params: dict, signature: str,
-    ) -> bool:
-        """Compute and compare a single Twilio signature."""
-        data_to_sign = url
-        for key in sorted(post_params.keys()):
-            data_to_sign += key + post_params[key]
-        mac = hmac.new(
-            self._auth_token.encode("utf-8"),
-            data_to_sign.encode("utf-8"),
-            hashlib.sha1,
-        )
-        computed = base64.b64encode(mac.digest()).decode("utf-8")
-        return hmac.compare_digest(computed, signature)
-
-    @staticmethod
-    def _port_variant_url(url: str) -> str | None:
-        """Return the URL with the default port toggled, or None.
-
-        Only toggles default ports (443 for https, 80 for http).
-        Non-standard ports are never modified.
-        """
-        parsed = urllib.parse.urlparse(url)
-        default_ports = {"https": 443, "http": 80}
-        default_port = default_ports.get(parsed.scheme)
-        if default_port is None:
-            return None
-
-        if parsed.port == default_port:
-            # Has explicit default port → strip it
-            return urllib.parse.urlunparse(
-                (parsed.scheme, parsed.hostname, parsed.path,
-                 parsed.params, parsed.query, parsed.fragment)
-            )
-        elif parsed.port is None:
-            # No port → add default
-            netloc = f"{parsed.hostname}:{default_port}"
-            return urllib.parse.urlunparse(
-                (parsed.scheme, netloc, parsed.path,
-                 parsed.params, parsed.query, parsed.fragment)
-            )
-
-        # Non-standard port — no variant
-        return None
+        content = re.sub(r"\*\*(.+?)\*\*", r"\1", content, flags=re.DOTALL)
+        content = re.sub(r"\*(.+?)\*", r"\1", content, flags=re.DOTALL)
+        content = re.sub(r"__(.+?)__", r"\1", content, flags=re.DOTALL)
+        content = re.sub(r"_(.+?)_", r"\1", content, flags=re.DOTALL)
+        content = re.sub(r"```[a-z]*\n?", "", content)
+        content = re.sub(r"`(.+?)`", r"\1", content)
+        content = re.sub(r"^#{1,6}\s+", "", content, flags=re.MULTILINE)
+        content = re.sub(r"\[([^\]]+)\]\([^\)]+\)", r"\1", content)
+        content = re.sub(r"\n{3,}", "\n\n", content)
+        return content.strip()

    # ------------------------------------------------------------------
    # Twilio webhook handler
@@ -289,7 +213,7 @@ class SmsAdapter(BasePlatformAdapter):
        try:
            raw = await request.read()
            # Twilio sends form-encoded data, not JSON
-            form = urllib.parse.parse_qs(raw.decode("utf-8"), keep_blank_values=True)
+            form = urllib.parse.parse_qs(raw.decode("utf-8"))
        except Exception as e:
            logger.error("[sms] webhook parse error: %s", e)
            return web.Response(
@@ -298,27 +222,6 @@ class SmsAdapter(BasePlatformAdapter):
                status=400,
            )

-        # Validate Twilio request signature when SMS_WEBHOOK_URL is configured
-        if self._webhook_url:
-            twilio_sig = request.headers.get("X-Twilio-Signature", "")
-            if not twilio_sig:
-                logger.warning("[sms] Rejected: missing X-Twilio-Signature header")
-                return web.Response(
-                    text='<?xml version="1.0" encoding="UTF-8"?><Response></Response>',
-                    content_type="application/xml",
-                    status=403,
-                )
-            flat_params = {k: v[0] for k, v in form.items() if v}
-            if not self._validate_twilio_signature(
-                self._webhook_url, flat_params, twilio_sig
-            ):
-                logger.warning("[sms] Rejected: invalid Twilio signature")
-                return web.Response(
-                    text='<?xml version="1.0" encoding="UTF-8"?><Response></Response>',
-                    content_type="application/xml",
-                    status=403,
-                )
-
        # Extract fields (parse_qs returns lists)
        from_number = (form.get("From", [""]))[0].strip()
        to_number = (form.get("To", [""]))[0].strip()
@@ -333,7 +236,7 @@ class SmsAdapter(BasePlatformAdapter):

        # Ignore messages from our own number (echo prevention)
        if from_number == self._from_number:
-            logger.debug("[sms] ignoring echo from own number %s", redact_phone(from_number))
+            logger.debug("[sms] ignoring echo from own number %s", _redact_phone(from_number))
            return web.Response(
                text='<?xml version="1.0" encoding="UTF-8"?><Response></Response>',
                content_type="application/xml",
@@ -341,8 +244,8 @@ class SmsAdapter(BasePlatformAdapter):

        logger.info(
            "[sms] inbound from %s -> %s: %s",
-            redact_phone(from_number),
-            redact_phone(to_number),
+            _redact_phone(from_number),
+            _redact_phone(to_number),
            text[:80],
        )

@@ -147,6 +147,7 @@ class TelegramAdapter(BasePlatformAdapter):
        self._text_batch_split_delay_seconds = float(os.getenv("HERMES_TELEGRAM_TEXT_BATCH_SPLIT_DELAY_SECONDS", "2.0"))
        self._pending_text_batches: Dict[str, MessageEvent] = {}
        self._pending_text_batch_tasks: Dict[str, asyncio.Task] = {}
+        self._token_lock_identity: Optional[str] = None
        self._polling_error_task: Optional[asyncio.Task] = None
        self._polling_conflict_count: int = 0
        self._polling_network_error_count: int = 0
@@ -299,11 +300,9 @@ class TelegramAdapter(BasePlatformAdapter):

        # Exhausted retries — fatal
        message = (
-            "Another process is already polling this Telegram bot token "
-            "(possibly OpenClaw or another Hermes instance). "
+            "Another Telegram bot poller is already using this token. "
            "Hermes stopped Telegram polling after %d retries. "
-            "Only one poller can run per token — stop the other process "
-            "and restart with 'hermes start'."
+            "Make sure only one gateway instance is running for this bot token."
            % MAX_CONFLICT_RETRIES
        )
        logger.error("[%s] %s Original error: %s", self.name, message, error)
@@ -498,7 +497,23 @@ class TelegramAdapter(BasePlatformAdapter):
            return False
        
        try:
-            if not self._acquire_platform_lock('telegram-bot-token', self.config.token, 'Telegram bot token'):
+            from gateway.status import acquire_scoped_lock
+
+            self._token_lock_identity = self.config.token
+            acquired, existing = acquire_scoped_lock(
+                "telegram-bot-token",
+                self._token_lock_identity,
+                metadata={"platform": self.platform.value},
+            )
+            if not acquired:
+                owner_pid = existing.get("pid") if isinstance(existing, dict) else None
+                message = (
+                    "Another local Hermes gateway is already using this Telegram bot token"
+                    + (f" (PID {owner_pid})." if owner_pid else ".")
+                    + " Stop the other gateway before starting a second Telegram poller."
+                )
+                logger.error("[%s] %s", self.name, message)
+                self._set_fatal_error("telegram_token_lock", message, retryable=False)
                return False

            # Build the application
@@ -722,7 +737,12 @@ class TelegramAdapter(BasePlatformAdapter):
            return True
            
        except Exception as e:
-            self._release_platform_lock()
+            if self._token_lock_identity:
+                try:
+                    from gateway.status import release_scoped_lock
+                    release_scoped_lock("telegram-bot-token", self._token_lock_identity)
+                except Exception:
+                    pass
            message = f"Telegram startup failed: {e}"
            self._set_fatal_error("telegram_connect_error", message, retryable=True)
            logger.error("[%s] Failed to connect to Telegram: %s", self.name, e, exc_info=True)
@@ -748,7 +768,12 @@ class TelegramAdapter(BasePlatformAdapter):
                await self._app.shutdown()
            except Exception as e:
                logger.warning("[%s] Error during Telegram disconnect: %s", self.name, e, exc_info=True)
-        self._release_platform_lock()
+        if self._token_lock_identity:
+            try:
+                from gateway.status import release_scoped_lock
+                release_scoped_lock("telegram-bot-token", self._token_lock_identity)
+            except Exception as e:
+                logger.warning("[%s] Error releasing Telegram token lock: %s", self.name, e, exc_info=True)

        for task in self._pending_photo_batch_tasks.values():
            if task and not task.done():
@@ -759,6 +784,7 @@ class TelegramAdapter(BasePlatformAdapter):
        self._mark_disconnected()
        self._app = None
        self._bot = None
+        self._token_lock_identity = None
        logger.info("[%s] Disconnected from Telegram", self.name)

    def _should_thread_reply(self, reply_to: Optional[str], chunk_index: int) -> bool:
@@ -201,7 +201,6 @@ class WebhookAdapter(BasePlatformAdapter):
            "dingtalk",
            "feishu",
            "wecom",
-            "wecom_callback",
            "weixin",
            "bluebubbles",
        ):
@@ -59,7 +59,6 @@ except ImportError:
    httpx = None  # type: ignore[assignment]

 from gateway.config import Platform, PlatformConfig
-from gateway.platforms.helpers import MessageDeduplicator
 from gateway.platforms.base import (
    BasePlatformAdapter,
    MessageEvent,
@@ -93,6 +92,7 @@ REQUEST_TIMEOUT_SECONDS = 15.0
 HEARTBEAT_INTERVAL_SECONDS = 30.0
 RECONNECT_BACKOFF = [2, 5, 10, 30, 60]

+DEDUP_WINDOW_SECONDS = 300
 DEDUP_MAX_SIZE = 1000

 IMAGE_MAX_BYTES = 10 * 1024 * 1024
@@ -172,7 +172,7 @@ class WeComAdapter(BasePlatformAdapter):
        self._listen_task: Optional[asyncio.Task] = None
        self._heartbeat_task: Optional[asyncio.Task] = None
        self._pending_responses: Dict[str, asyncio.Future] = {}
-        self._dedup = MessageDeduplicator(max_size=DEDUP_MAX_SIZE)
+        self._seen_messages: Dict[str, float] = {}
        self._reply_req_ids: Dict[str, str] = {}

        # Text batching: merge rapid successive messages (Telegram-style).
@@ -250,7 +250,7 @@ class WeComAdapter(BasePlatformAdapter):
            await self._http_client.aclose()
            self._http_client = None

-        self._dedup.clear()
+        self._seen_messages.clear()
        logger.info("[%s] Disconnected", self.name)

    async def _cleanup_ws(self) -> None:
@@ -476,7 +476,7 @@ class WeComAdapter(BasePlatformAdapter):
            return

        msg_id = str(body.get("msgid") or self._payload_req_id(payload) or uuid.uuid4().hex)
-        if self._dedup.is_duplicate(msg_id):
+        if self._is_duplicate(msg_id):
            logger.debug("[%s] Duplicate message %s ignored", self.name, msg_id)
            return
        self._remember_reply_req_id(msg_id, self._payload_req_id(payload))
@@ -636,13 +636,6 @@ class WeComAdapter(BasePlatformAdapter):
                if voice_text:
                    text_parts.append(voice_text)

-            # Extract appmsg title (filename) for WeCom AI Bot attachments
-            if msgtype == "appmsg":
-                appmsg = body.get("appmsg") if isinstance(body.get("appmsg"), dict) else {}
-                title = str(appmsg.get("title") or "").strip()
-                if title:
-                    text_parts.append(title)
-
        quote = body.get("quote") if isinstance(body.get("quote"), dict) else {}
        quote_type = str(quote.get("msgtype") or "").lower()
        if quote_type == "text":
@@ -675,13 +668,6 @@ class WeComAdapter(BasePlatformAdapter):
                refs.append(("image", body["image"]))
            if msgtype == "file" and isinstance(body.get("file"), dict):
                refs.append(("file", body["file"]))
-            # Handle appmsg (WeCom AI Bot attachments with PDF/Word/Excel)
-            if msgtype == "appmsg" and isinstance(body.get("appmsg"), dict):
-                appmsg = body["appmsg"]
-                if isinstance(appmsg.get("file"), dict):
-                    refs.append(("file", appmsg["file"]))
-                elif isinstance(appmsg.get("image"), dict):
-                    refs.append(("image", appmsg["image"]))

        quote = body.get("quote") if isinstance(body.get("quote"), dict) else {}
        quote_type = str(quote.get("msgtype") or "").lower()
@@ -839,6 +825,24 @@ class WeComAdapter(BasePlatformAdapter):
        wildcard = self._groups.get("*")
        return wildcard if isinstance(wildcard, dict) else {}

+    def _is_duplicate(self, msg_id: str) -> bool:
+        now = time.time()
+        if len(self._seen_messages) > DEDUP_MAX_SIZE:
+            cutoff = now - DEDUP_WINDOW_SECONDS
+            self._seen_messages = {
+                key: ts for key, ts in self._seen_messages.items() if ts > cutoff
+            }
+            if self._reply_req_ids:
+                self._reply_req_ids = {
+                    key: value for key, value in self._reply_req_ids.items() if key in self._seen_messages
+                }
+
+        if msg_id in self._seen_messages:
+            return True
+
+        self._seen_messages[msg_id] = now
+        return False
+
    def _remember_reply_req_id(self, message_id: str, req_id: str) -> None:
        normalized_message_id = str(message_id or "").strip()
        normalized_req_id = str(req_id or "").strip()
@@ -1,387 +0,0 @@
-"""WeCom callback-mode adapter for self-built enterprise applications.
-
-Unlike the bot/websocket adapter in ``wecom.py``, this handles the standard
-WeCom callback flow: WeCom POSTs encrypted XML to an HTTP endpoint, the
-adapter decrypts it, queues the message for the agent, and immediately
-acknowledges.  The agent's reply is delivered later via the proactive
-``message/send`` API using an access-token.
-
-Supports multiple self-built apps under one gateway instance, scoped by
-``corp_id:user_id`` to avoid cross-corp collisions.
-"""
-
-from __future__ import annotations
-
-import asyncio
-import logging
-import socket as _socket
-import time
-from typing import Any, Dict, List, Optional
-from xml.etree import ElementTree as ET
-
-try:
-    from aiohttp import web
-
-    AIOHTTP_AVAILABLE = True
-except ImportError:
-    web = None  # type: ignore[assignment]
-    AIOHTTP_AVAILABLE = False
-
-try:
-    import httpx
-
-    HTTPX_AVAILABLE = True
-except ImportError:
-    httpx = None  # type: ignore[assignment]
-    HTTPX_AVAILABLE = False
-
-from gateway.config import Platform, PlatformConfig
-from gateway.platforms.base import BasePlatformAdapter, MessageEvent, MessageType, SendResult
-from gateway.platforms.wecom_crypto import WXBizMsgCrypt, WeComCryptoError
-
-logger = logging.getLogger(__name__)
-
-DEFAULT_HOST = "0.0.0.0"
-DEFAULT_PORT = 8645
-DEFAULT_PATH = "/wecom/callback"
-ACCESS_TOKEN_TTL_SECONDS = 7200
-MESSAGE_DEDUP_TTL_SECONDS = 300
-
-
-def check_wecom_callback_requirements() -> bool:
-    return AIOHTTP_AVAILABLE and HTTPX_AVAILABLE
-
-
-class WecomCallbackAdapter(BasePlatformAdapter):
-    def __init__(self, config: PlatformConfig):
-        super().__init__(config, Platform.WECOM_CALLBACK)
-        extra = config.extra or {}
-        self._host = str(extra.get("host") or DEFAULT_HOST)
-        self._port = int(extra.get("port") or DEFAULT_PORT)
-        self._path = str(extra.get("path") or DEFAULT_PATH)
-        self._apps: List[Dict[str, Any]] = self._normalize_apps(extra)
-        self._runner: Optional[web.AppRunner] = None
-        self._site: Optional[web.TCPSite] = None
-        self._app: Optional[web.Application] = None
-        self._http_client: Optional[httpx.AsyncClient] = None
-        self._message_queue: asyncio.Queue[MessageEvent] = asyncio.Queue()
-        self._poll_task: Optional[asyncio.Task] = None
-        self._seen_messages: Dict[str, float] = {}
-        self._user_app_map: Dict[str, str] = {}
-        self._access_tokens: Dict[str, Dict[str, Any]] = {}
-
-    # ------------------------------------------------------------------
-    # App normalisation
-    # ------------------------------------------------------------------
-
-    @staticmethod
-    def _user_app_key(corp_id: str, user_id: str) -> str:
-        return f"{corp_id}:{user_id}" if corp_id else user_id
-
-    @staticmethod
-    def _normalize_apps(extra: Dict[str, Any]) -> List[Dict[str, Any]]:
-        apps = extra.get("apps")
-        if isinstance(apps, list) and apps:
-            return [dict(app) for app in apps if isinstance(app, dict)]
-        if extra.get("corp_id"):
-            return [
-                {
-                    "name": extra.get("name") or "default",
-                    "corp_id": extra.get("corp_id", ""),
-                    "corp_secret": extra.get("corp_secret", ""),
-                    "agent_id": str(extra.get("agent_id", "")),
-                    "token": extra.get("token", ""),
-                    "encoding_aes_key": extra.get("encoding_aes_key", ""),
-                }
-            ]
-        return []
-
-    # ------------------------------------------------------------------
-    # Lifecycle
-    # ------------------------------------------------------------------
-
-    async def connect(self) -> bool:
-        if not self._apps:
-            logger.warning("[WecomCallback] No callback apps configured")
-            return False
-        if not check_wecom_callback_requirements():
-            logger.warning("[WecomCallback] aiohttp/httpx not installed")
-            return False
-
-        # Quick port-in-use check.
-        try:
-            with _socket.socket(_socket.AF_INET, _socket.SOCK_STREAM) as sock:
-                sock.settimeout(1)
-                sock.connect(("127.0.0.1", self._port))
-            logger.error("[WecomCallback] Port %d already in use", self._port)
-            return False
-        except (ConnectionRefusedError, OSError):
-            pass
-
-        try:
-            self._http_client = httpx.AsyncClient(timeout=20.0)
-            self._app = web.Application()
-            self._app.router.add_get("/health", self._handle_health)
-            self._app.router.add_get(self._path, self._handle_verify)
-            self._app.router.add_post(self._path, self._handle_callback)
-            self._runner = web.AppRunner(self._app)
-            await self._runner.setup()
-            self._site = web.TCPSite(self._runner, self._host, self._port)
-            await self._site.start()
-            self._poll_task = asyncio.create_task(self._poll_loop())
-            self._mark_connected()
-            logger.info(
-                "[WecomCallback] HTTP server listening on %s:%s%s",
-                self._host, self._port, self._path,
-            )
-            for app in self._apps:
-                try:
-                    await self._refresh_access_token(app)
-                except Exception as exc:
-                    logger.warning(
-                        "[WecomCallback] Initial token refresh failed for app '%s': %s",
-                        app.get("name", "default"), exc,
-                    )
-            return True
-        except Exception:
-            await self._cleanup()
-            logger.exception("[WecomCallback] Failed to start")
-            return False
-
-    async def disconnect(self) -> None:
-        self._running = False
-        if self._poll_task:
-            self._poll_task.cancel()
-            try:
-                await self._poll_task
-            except asyncio.CancelledError:
-                pass
-            self._poll_task = None
-        await self._cleanup()
-        self._mark_disconnected()
-        logger.info("[WecomCallback] Disconnected")
-
-    async def _cleanup(self) -> None:
-        self._site = None
-        if self._runner:
-            await self._runner.cleanup()
-            self._runner = None
-        self._app = None
-        if self._http_client:
-            await self._http_client.aclose()
-            self._http_client = None
-
-    # ------------------------------------------------------------------
-    # Outbound: proactive send via access-token API
-    # ------------------------------------------------------------------
-
-    async def send(
-        self,
-        chat_id: str,
-        content: str,
-        reply_to: Optional[str] = None,
-        metadata: Optional[Dict[str, Any]] = None,
-    ) -> SendResult:
-        app = self._resolve_app_for_chat(chat_id)
-        touser = chat_id.split(":", 1)[1] if ":" in chat_id else chat_id
-        try:
-            token = await self._get_access_token(app)
-            payload = {
-                "touser": touser,
-                "msgtype": "text",
-                "agentid": int(str(app.get("agent_id") or 0)),
-                "text": {"content": content[:2048]},
-                "safe": 0,
-            }
-            resp = await self._http_client.post(
-                f"https://qyapi.weixin.qq.com/cgi-bin/message/send?access_token={token}",
-                json=payload,
-            )
-            data = resp.json()
-            if data.get("errcode") != 0:
-                return SendResult(success=False, error=str(data))
-            return SendResult(
-                success=True,
-                message_id=str(data.get("msgid", "")),
-                raw_response=data,
-            )
-        except Exception as exc:
-            return SendResult(success=False, error=str(exc))
-
-    def _resolve_app_for_chat(self, chat_id: str) -> Dict[str, Any]:
-        """Pick the app associated with *chat_id*, falling back sensibly."""
-        app_name = self._user_app_map.get(chat_id)
-        if not app_name and ":" not in chat_id:
-            # Legacy bare user_id — try to find a unique match.
-            matching = [k for k in self._user_app_map if k.endswith(f":{chat_id}")]
-            if len(matching) == 1:
-                app_name = self._user_app_map.get(matching[0])
-        app = self._get_app_by_name(app_name) if app_name else None
-        return app or self._apps[0]
-
-    async def get_chat_info(self, chat_id: str) -> Dict[str, Any]:
-        return {"name": chat_id, "type": "dm"}
-
-    # ------------------------------------------------------------------
-    # Inbound: HTTP callback handlers
-    # ------------------------------------------------------------------
-
-    async def _handle_health(self, request: web.Request) -> web.Response:
-        return web.json_response({"status": "ok", "platform": "wecom_callback"})
-
-    async def _handle_verify(self, request: web.Request) -> web.Response:
-        """GET endpoint — WeCom URL verification handshake."""
-        msg_signature = request.query.get("msg_signature", "")
-        timestamp = request.query.get("timestamp", "")
-        nonce = request.query.get("nonce", "")
-        echostr = request.query.get("echostr", "")
-        for app in self._apps:
-            try:
-                crypt = self._crypt_for_app(app)
-                plain = crypt.verify_url(msg_signature, timestamp, nonce, echostr)
-                return web.Response(text=plain, content_type="text/plain")
-            except Exception:
-                continue
-        return web.Response(status=403, text="signature verification failed")
-
-    async def _handle_callback(self, request: web.Request) -> web.Response:
-        """POST endpoint — receive an encrypted message callback."""
-        msg_signature = request.query.get("msg_signature", "")
-        timestamp = request.query.get("timestamp", "")
-        nonce = request.query.get("nonce", "")
-        body = await request.text()
-
-        for app in self._apps:
-            try:
-                decrypted = self._decrypt_request(
-                    app, body, msg_signature, timestamp, nonce,
-                )
-                event = self._build_event(app, decrypted)
-                if event is not None:
-                    # Record which app this user belongs to.
-                    if event.source and event.source.user_id:
-                        map_key = self._user_app_key(
-                            str(app.get("corp_id") or ""), event.source.user_id,
-                        )
-                        self._user_app_map[map_key] = app["name"]
-                    await self._message_queue.put(event)
-                # Immediately acknowledge — the agent's reply will arrive
-                # later via the proactive message/send API.
-                return web.Response(text="success", content_type="text/plain")
-            except WeComCryptoError:
-                continue
-            except Exception:
-                logger.exception("[WecomCallback] Error handling message")
-                break
-        return web.Response(status=400, text="invalid callback payload")
-
-    async def _poll_loop(self) -> None:
-        """Drain the message queue and dispatch to the gateway runner."""
-        while True:
-            event = await self._message_queue.get()
-            try:
-                task = asyncio.create_task(self.handle_message(event))
-                self._background_tasks.add(task)
-                task.add_done_callback(self._background_tasks.discard)
-            except Exception:
-                logger.exception("[WecomCallback] Failed to enqueue event")
-
-    # ------------------------------------------------------------------
-    # XML / crypto helpers
-    # ------------------------------------------------------------------
-
-    def _decrypt_request(
-        self, app: Dict[str, Any], body: str,
-        msg_signature: str, timestamp: str, nonce: str,
-    ) -> str:
-        root = ET.fromstring(body)
-        encrypt = root.findtext("Encrypt", default="")
-        crypt = self._crypt_for_app(app)
-        return crypt.decrypt(msg_signature, timestamp, nonce, encrypt).decode("utf-8")
-
-    def _build_event(self, app: Dict[str, Any], xml_text: str) -> Optional[MessageEvent]:
-        root = ET.fromstring(xml_text)
-        msg_type = (root.findtext("MsgType") or "").lower()
-        # Silently acknowledge lifecycle events.
-        if msg_type == "event":
-            event_name = (root.findtext("Event") or "").lower()
-            if event_name in {"enter_agent", "subscribe"}:
-                return None
-        if msg_type not in {"text", "event"}:
-            return None
-
-        user_id = root.findtext("FromUserName", default="")
-        corp_id = root.findtext("ToUserName", default=app.get("corp_id", ""))
-        scoped_chat_id = self._user_app_key(corp_id, user_id)
-        content = root.findtext("Content", default="").strip()
-        if not content and msg_type == "event":
-            content = "/start"
-        msg_id = (
-            root.findtext("MsgId")
-            or f"{user_id}:{root.findtext('CreateTime', default='0')}"
-        )
-        source = self.build_source(
-            chat_id=scoped_chat_id,
-            chat_name=user_id,
-            chat_type="dm",
-            user_id=user_id,
-            user_name=user_id,
-        )
-        return MessageEvent(
-            text=content,
-            message_type=MessageType.TEXT,
-            source=source,
-            raw_message=xml_text,
-            message_id=msg_id,
-        )
-
-    def _crypt_for_app(self, app: Dict[str, Any]) -> WXBizMsgCrypt:
-        return WXBizMsgCrypt(
-            token=str(app.get("token") or ""),
-            encoding_aes_key=str(app.get("encoding_aes_key") or ""),
-            receive_id=str(app.get("corp_id") or ""),
-        )
-
-    def _get_app_by_name(self, name: Optional[str]) -> Optional[Dict[str, Any]]:
-        if not name:
-            return None
-        for app in self._apps:
-            if app.get("name") == name:
-                return app
-        return None
-
-    # ------------------------------------------------------------------
-    # Access-token management
-    # ------------------------------------------------------------------
-
-    async def _get_access_token(self, app: Dict[str, Any]) -> str:
-        cached = self._access_tokens.get(app["name"])
-        now = time.time()
-        if cached and cached.get("expires_at", 0) > now + 60:
-            return cached["token"]
-        return await self._refresh_access_token(app)
-
-    async def _refresh_access_token(self, app: Dict[str, Any]) -> str:
-        resp = await self._http_client.get(
-            "https://qyapi.weixin.qq.com/cgi-bin/gettoken",
-            params={
-                "corpid": app.get("corp_id"),
-                "corpsecret": app.get("corp_secret"),
-            },
-        )
-        data = resp.json()
-        if data.get("errcode") != 0:
-            raise RuntimeError(f"WeCom token refresh failed: {data}")
-        token = data["access_token"]
-        expires_in = int(data.get("expires_in", ACCESS_TOKEN_TTL_SECONDS))
-        self._access_tokens[app["name"]] = {
-            "token": token,
-            "expires_at": time.time() + expires_in,
-        }
-        logger.info(
-            "[WecomCallback] Token refreshed for app '%s' (corp=%s), expires in %ss",
-            app.get("name", "default"),
-            app.get("corp_id", ""),
-            expires_in,
-        )
-        return token
@@ -1,142 +0,0 @@
-"""WeCom BizMsgCrypt-compatible AES-CBC encryption for callback mode.
-
-Implements the same wire format as Tencent's official ``WXBizMsgCrypt``
-SDK so that WeCom can verify, encrypt, and decrypt callback payloads.
-"""
-
-from __future__ import annotations
-
-import base64
-import hashlib
-import os
-import secrets
-import socket
-import struct
-from typing import Optional
-from xml.etree import ElementTree as ET
-
-from cryptography.hazmat.backends import default_backend
-from cryptography.hazmat.primitives.ciphers import Cipher, algorithms, modes
-
-
-class WeComCryptoError(Exception):
-    pass
-
-
-class SignatureError(WeComCryptoError):
-    pass
-
-
-class DecryptError(WeComCryptoError):
-    pass
-
-
-class EncryptError(WeComCryptoError):
-    pass
-
-
-class PKCS7Encoder:
-    block_size = 32
-
-    @classmethod
-    def encode(cls, text: bytes) -> bytes:
-        amount_to_pad = cls.block_size - (len(text) % cls.block_size)
-        if amount_to_pad == 0:
-            amount_to_pad = cls.block_size
-        pad = bytes([amount_to_pad]) * amount_to_pad
-        return text + pad
-
-    @classmethod
-    def decode(cls, decrypted: bytes) -> bytes:
-        if not decrypted:
-            raise DecryptError("empty decrypted payload")
-        pad = decrypted[-1]
-        if pad < 1 or pad > cls.block_size:
-            raise DecryptError("invalid PKCS7 padding")
-        if decrypted[-pad:] != bytes([pad]) * pad:
-            raise DecryptError("malformed PKCS7 padding")
-        return decrypted[:-pad]
-
-
-def _sha1_signature(token: str, timestamp: str, nonce: str, encrypt: str) -> str:
-    parts = sorted([token, timestamp, nonce, encrypt])
-    return hashlib.sha1("".join(parts).encode("utf-8")).hexdigest()
-
-
-class WXBizMsgCrypt:
-    """Minimal WeCom callback crypto helper compatible with BizMsgCrypt semantics."""
-
-    def __init__(self, token: str, encoding_aes_key: str, receive_id: str):
-        if not token:
-            raise ValueError("token is required")
-        if not encoding_aes_key:
-            raise ValueError("encoding_aes_key is required")
-        if len(encoding_aes_key) != 43:
-            raise ValueError("encoding_aes_key must be 43 chars")
-        if not receive_id:
-            raise ValueError("receive_id is required")
-
-        self.token = token
-        self.receive_id = receive_id
-        self.key = base64.b64decode(encoding_aes_key + "=")
-        self.iv = self.key[:16]
-
-    def verify_url(self, msg_signature: str, timestamp: str, nonce: str, echostr: str) -> str:
-        plain = self.decrypt(msg_signature, timestamp, nonce, echostr)
-        return plain.decode("utf-8")
-
-    def decrypt(self, msg_signature: str, timestamp: str, nonce: str, encrypt: str) -> bytes:
-        expected = _sha1_signature(self.token, timestamp, nonce, encrypt)
-        if expected != msg_signature:
-            raise SignatureError("signature mismatch")
-        try:
-            cipher_text = base64.b64decode(encrypt)
-        except Exception as exc:
-            raise DecryptError(f"invalid base64 payload: {exc}") from exc
-        try:
-            cipher = Cipher(algorithms.AES(self.key), modes.CBC(self.iv), backend=default_backend())
-            decryptor = cipher.decryptor()
-            padded = decryptor.update(cipher_text) + decryptor.finalize()
-            plain = PKCS7Encoder.decode(padded)
-            content = plain[16:]  # skip 16-byte random prefix
-            xml_length = socket.ntohl(struct.unpack("I", content[:4])[0])
-            xml_content = content[4:4 + xml_length]
-            receive_id = content[4 + xml_length:].decode("utf-8")
-        except WeComCryptoError:
-            raise
-        except Exception as exc:
-            raise DecryptError(f"decrypt failed: {exc}") from exc
-
-        if receive_id != self.receive_id:
-            raise DecryptError("receive_id mismatch")
-        return xml_content
-
-    def encrypt(self, plaintext: str, nonce: Optional[str] = None, timestamp: Optional[str] = None) -> str:
-        nonce = nonce or self._random_nonce()
-        timestamp = timestamp or str(int(__import__("time").time()))
-        encrypt = self._encrypt_bytes(plaintext.encode("utf-8"))
-        signature = _sha1_signature(self.token, timestamp, nonce, encrypt)
-        root = ET.Element("xml")
-        ET.SubElement(root, "Encrypt").text = encrypt
-        ET.SubElement(root, "MsgSignature").text = signature
-        ET.SubElement(root, "TimeStamp").text = timestamp
-        ET.SubElement(root, "Nonce").text = nonce
-        return ET.tostring(root, encoding="unicode")
-
-    def _encrypt_bytes(self, raw: bytes) -> str:
-        try:
-            random_prefix = os.urandom(16)
-            msg_len = struct.pack("I", socket.htonl(len(raw)))
-            payload = random_prefix + msg_len + raw + self.receive_id.encode("utf-8")
-            padded = PKCS7Encoder.encode(payload)
-            cipher = Cipher(algorithms.AES(self.key), modes.CBC(self.iv), backend=default_backend())
-            encryptor = cipher.encryptor()
-            encrypted = encryptor.update(padded) + encryptor.finalize()
-            return base64.b64encode(encrypted).decode("utf-8")
-        except Exception as exc:
-            raise EncryptError(f"encrypt failed: {exc}") from exc
-
-    @staticmethod
-    def _random_nonce(length: int = 10) -> str:
-        alphabet = "0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ"
-        return "".join(secrets.choice(alphabet) for _ in range(length))
@@ -53,7 +53,6 @@ except ImportError:  # pragma: no cover - dependency gate
    CRYPTO_AVAILABLE = False

 from gateway.config import Platform, PlatformConfig
-from gateway.platforms.helpers import MessageDeduplicator
 from gateway.platforms.base import (
    BasePlatformAdapter,
    MessageEvent,
@@ -64,7 +63,6 @@ from gateway.platforms.base import (
    cache_image_from_bytes,
 )
 from hermes_constants import get_hermes_home
-from utils import atomic_json_write

 ILINK_BASE_URL = "https://ilinkai.weixin.qq.com"
 WEIXIN_CDN_BASE_URL = "https://novac2c.cdn.weixin.qq.com/c2c"
@@ -208,7 +206,7 @@ def save_weixin_account(
        "saved_at": time.strftime("%Y-%m-%dT%H:%M:%SZ", time.gmtime()),
    }
    path = _account_file(hermes_home, account_id)
-    atomic_json_write(path, payload)
+    path.write_text(json.dumps(payload, indent=2), encoding="utf-8")
    try:
        path.chmod(0o600)
    except OSError:
@@ -271,7 +269,7 @@ class ContextTokenStore:
            if key.startswith(prefix)
        }
        try:
-            atomic_json_write(self._path(account_id), payload)
+            self._path(account_id).write_text(json.dumps(payload), encoding="utf-8")
        except Exception as exc:
            logger.warning("weixin: failed to persist context tokens for %s: %s", _safe_id(account_id), exc)

@@ -734,42 +732,6 @@ def _split_delivery_units_for_weixin(content: str) -> List[str]:
    return [unit for unit in units if unit]


-def _looks_like_chatty_line_for_weixin(line: str) -> bool:
-    """Return True when a line looks like a standalone chat utterance."""
-    stripped = line.strip()
-    if not stripped:
-        return False
-    if len(stripped) > 48:
-        return False
-    if line.startswith((" ", "\t")):
-        return False
-    if stripped.startswith((">", "-", "*", "【")):
-        return False
-    if re.match(r"^\*\*[^*]+\*\*$", stripped):
-        return False
-    if re.match(r"^\d+\.\s", stripped):
-        return False
-    return True
-
-
-def _looks_like_heading_line_for_weixin(line: str) -> bool:
-    """Return True when a short line behaves like a plain-text heading."""
-    stripped = line.strip()
-    if not stripped:
-        return False
-    return len(stripped) <= 24 and stripped.endswith((":", "："))
-
-
-def _should_split_short_chat_block_for_weixin(block: str) -> bool:
-    """Split only chat-like multiline blocks into separate bubbles."""
-    lines = [line for line in block.splitlines() if line.strip()]
-    if not 2 <= len(lines) <= 6:
-        return False
-    if _looks_like_heading_line_for_weixin(lines[0]):
-        return False
-    return all(_looks_like_chatty_line_for_weixin(line) for line in lines)
-
-
 def _pack_markdown_blocks_for_weixin(content: str, max_length: int) -> List[str]:
    if len(content) <= max_length:
        return [content]
@@ -793,64 +755,23 @@ def _pack_markdown_blocks_for_weixin(content: str, max_length: int) -> List[str]
    return packed


-def _split_text_for_weixin_delivery(
-    content: str, max_length: int, split_per_line: bool = False,
-) -> List[str]:
+def _split_text_for_weixin_delivery(content: str, max_length: int) -> List[str]:
    """Split content into sequential Weixin messages.

-    *compact* (default): Keep everything in a single message whenever it fits
-    within the platform limit, even when the author used explicit line breaks.
-    Only fall back to block-aware packing when the payload exceeds
-    ``max_length``.
-
-    *per_line* (``split_per_line=True``): Legacy behavior — top-level line
-    breaks become separate chat messages; oversized units still use
-    block-aware packing.
-
-    The active mode is controlled via ``config.yaml`` ->
-    ``platforms.weixin.extra.split_multiline_messages`` (``true`` / ``false``)
-    or the env var ``WEIXIN_SPLIT_MULTILINE_MESSAGES``.
+    Prefer one message per top-level line/markdown unit when the author used
+    explicit line breaks. Oversized units fall back to block-aware packing so
+    long code fences still split safely.
    """
-    if split_per_line:
-        # Legacy: one message per top-level delivery unit.
-        if len(content) <= max_length and "\n" not in content:
-            return [content]
-        chunks: List[str] = []
-        for unit in _split_delivery_units_for_weixin(content):
-            if len(unit) <= max_length:
-                chunks.append(unit)
-                continue
-            chunks.extend(_pack_markdown_blocks_for_weixin(unit, max_length))
-        return chunks or [content]
+    if len(content) <= max_length and "\n" not in content:
+        return [content]

-    # Compact (default): single message when under the limit — unless the
-    # content looks like a short chatty exchange, in which case split into
-    # separate bubbles for a more natural chat feel.
-    if len(content) <= max_length:
-        return (
-            _split_delivery_units_for_weixin(content)
-            if _should_split_short_chat_block_for_weixin(content)
-            else [content]
-        )
-    return _pack_markdown_blocks_for_weixin(content, max_length) or [content]
-
-
-def _coerce_bool(value: Any, default: bool = True) -> bool:
-    """Coerce a config value to bool, tolerating strings like ``"true"``."""
-    if value is None:
-        return default
-    if isinstance(value, bool):
-        return value
-    if isinstance(value, (int, float)):
-        return bool(value)
-    text = str(value).strip().lower()
-    if not text:
-        return default
-    if text in {"1", "true", "yes", "on"}:
-        return True
-    if text in {"0", "false", "no", "off"}:
-        return False
-    return default
+    chunks: List[str] = []
+    for unit in _split_delivery_units_for_weixin(content):
+        if len(unit) <= max_length:
+            chunks.append(unit)
+            continue
+        chunks.extend(_pack_markdown_blocks_for_weixin(unit, max_length))
+    return chunks or [content]


 def _extract_text(item_list: List[Dict[str, Any]]) -> str:
@@ -912,7 +833,7 @@ def _load_sync_buf(hermes_home: str, account_id: str) -> str:

 def _save_sync_buf(hermes_home: str, account_id: str, sync_buf: str) -> None:
    path = _sync_buf_path(hermes_home, account_id)
-    atomic_json_write(path, {"get_updates_buf": sync_buf})
+    path.write_text(json.dumps({"get_updates_buf": sync_buf}), encoding="utf-8")


 async def qr_login(
@@ -1051,7 +972,8 @@ class WeixinAdapter(BasePlatformAdapter):
        self._typing_cache = TypingTicketCache()
        self._session: Optional[aiohttp.ClientSession] = None
        self._poll_task: Optional[asyncio.Task] = None
-        self._dedup = MessageDeduplicator(ttl_seconds=MESSAGE_DEDUP_TTL_SECONDS)
+        self._seen_messages: Dict[str, float] = {}
+        self._token_lock_identity: Optional[str] = None

        self._account_id = str(extra.get("account_id") or os.getenv("WEIXIN_ACCOUNT_ID", "")).strip()
        self._token = str(config.token or extra.get("token") or os.getenv("WEIXIN_TOKEN", "")).strip()
@@ -1059,16 +981,6 @@ class WeixinAdapter(BasePlatformAdapter):
        self._cdn_base_url = str(
            extra.get("cdn_base_url") or os.getenv("WEIXIN_CDN_BASE_URL", WEIXIN_CDN_BASE_URL)
        ).strip().rstrip("/")
-        self._send_chunk_delay_seconds = float(
-            extra.get("send_chunk_delay_seconds") or os.getenv("WEIXIN_SEND_CHUNK_DELAY_SECONDS", "0.35")
-        )
-        self._send_chunk_retries = int(
-            extra.get("send_chunk_retries") or os.getenv("WEIXIN_SEND_CHUNK_RETRIES", "2")
-        )
-        self._send_chunk_retry_delay_seconds = float(
-            extra.get("send_chunk_retry_delay_seconds")
-            or os.getenv("WEIXIN_SEND_CHUNK_RETRY_DELAY_SECONDS", "1.0")
-        )
        self._dm_policy = str(extra.get("dm_policy") or os.getenv("WEIXIN_DM_POLICY", "open")).strip().lower()
        self._group_policy = str(extra.get("group_policy") or os.getenv("WEIXIN_GROUP_POLICY", "disabled")).strip().lower()
        allow_from = extra.get("allow_from")
@@ -1079,11 +991,6 @@ class WeixinAdapter(BasePlatformAdapter):
            group_allow_from = os.getenv("WEIXIN_GROUP_ALLOWED_USERS", "")
        self._allow_from = self._coerce_list(allow_from)
        self._group_allow_from = self._coerce_list(group_allow_from)
-        self._split_multiline_messages = _coerce_bool(
-            extra.get("split_multiline_messages")
-            or os.getenv("WEIXIN_SPLIT_MULTILINE_MESSAGES"),
-            default=False,
-        )

        if self._account_id and not self._token:
            persisted = load_weixin_account(hermes_home, self._account_id)
@@ -1119,7 +1026,23 @@ class WeixinAdapter(BasePlatformAdapter):
            return False

        try:
-            if not self._acquire_platform_lock('weixin-bot-token', self._token, 'Weixin bot token'):
+            from gateway.status import acquire_scoped_lock
+
+            self._token_lock_identity = self._token
+            acquired, existing = acquire_scoped_lock(
+                "weixin-bot-token",
+                self._token_lock_identity,
+                metadata={"platform": self.platform.value},
+            )
+            if not acquired:
+                owner_pid = existing.get("pid") if isinstance(existing, dict) else None
+                message = (
+                    "Another local Hermes gateway is already using this Weixin token"
+                    + (f" (PID {owner_pid})." if owner_pid else ".")
+                    + " Stop the other gateway before starting a second Weixin poller."
+                )
+                logger.error("[%s] %s", self.name, message)
+                self._set_fatal_error("weixin_token_lock", message, retryable=False)
                return False
        except Exception as exc:
            logger.debug("[%s] Token lock unavailable (non-fatal): %s", self.name, exc)
@@ -1143,7 +1066,12 @@ class WeixinAdapter(BasePlatformAdapter):
        if self._session and not self._session.closed:
            await self._session.close()
        self._session = None
-        self._release_platform_lock()
+        if self._token_lock_identity:
+            try:
+                from gateway.status import release_scoped_lock
+                release_scoped_lock("weixin-bot-token", self._token_lock_identity)
+            except Exception as exc:
+                logger.warning("[%s] Error releasing Weixin token lock: %s", self.name, exc, exc_info=True)
        self._mark_disconnected()
        logger.info("[%s] Disconnected", self.name)

@@ -1221,8 +1149,16 @@ class WeixinAdapter(BasePlatformAdapter):
            return

        message_id = str(message.get("message_id") or "").strip()
-        if message_id and self._dedup.is_duplicate(message_id):
-            return
+        if message_id:
+            now = time.time()
+            self._seen_messages = {
+                key: value
+                for key, value in self._seen_messages.items()
+                if now - value < MESSAGE_DEDUP_TTL_SECONDS
+            }
+            if message_id in self._seen_messages:
+                return
+            self._seen_messages[message_id] = now

        chat_type, effective_chat_id = _guess_chat_type(message, self._account_id)
        if chat_type == "group":
@@ -1394,50 +1330,7 @@ class WeixinAdapter(BasePlatformAdapter):
            logger.debug("[%s] getConfig failed for %s: %s", self.name, _safe_id(user_id), exc)

    def _split_text(self, content: str) -> List[str]:
-        return _split_text_for_weixin_delivery(
-            content, self.MAX_MESSAGE_LENGTH, self._split_multiline_messages,
-        )
-
-    async def _send_text_chunk(
-        self,
-        *,
-        chat_id: str,
-        chunk: str,
-        context_token: Optional[str],
-        client_id: str,
-    ) -> None:
-        """Send a single text chunk with per-chunk retry and backoff."""
-        last_error: Optional[Exception] = None
-        for attempt in range(self._send_chunk_retries + 1):
-            try:
-                await _send_message(
-                    self._session,
-                    base_url=self._base_url,
-                    token=self._token,
-                    to=chat_id,
-                    text=chunk,
-                    context_token=context_token,
-                    client_id=client_id,
-                )
-                return
-            except Exception as exc:
-                last_error = exc
-                if attempt >= self._send_chunk_retries:
-                    break
-                wait = self._send_chunk_retry_delay_seconds * (attempt + 1)
-                logger.warning(
-                    "[%s] send chunk failed to=%s attempt=%d/%d, retrying in %.2fs: %s",
-                    self.name,
-                    _safe_id(chat_id),
-                    attempt + 1,
-                    self._send_chunk_retries + 1,
-                    wait,
-                    exc,
-                )
-                if wait > 0:
-                    await asyncio.sleep(wait)
-        assert last_error is not None
-        raise last_error
+        return _split_text_for_weixin_delivery(content, self.MAX_MESSAGE_LENGTH)

    async def send(
        self,
@@ -1451,18 +1344,18 @@ class WeixinAdapter(BasePlatformAdapter):
        context_token = self._token_store.get(self._account_id, chat_id)
        last_message_id: Optional[str] = None
        try:
-            chunks = self._split_text(self.format_message(content))
-            for idx, chunk in enumerate(chunks):
+            for chunk in self._split_text(self.format_message(content)):
                client_id = f"hermes-weixin-{uuid.uuid4().hex}"
-                await self._send_text_chunk(
-                    chat_id=chat_id,
-                    chunk=chunk,
+                await _send_message(
+                    self._session,
+                    base_url=self._base_url,
+                    token=self._token,
+                    to=chat_id,
+                    text=chunk,
                    context_token=context_token,
                    client_id=client_id,
                )
                last_message_id = client_id
-                if idx < len(chunks) - 1 and self._send_chunk_delay_seconds > 0:
-                    await asyncio.sleep(self._send_chunk_delay_seconds)
            return SendResult(success=True, message_id=last_message_id)
        except Exception as exc:
            logger.error("[%s] send failed to=%s: %s", self.name, _safe_id(chat_id), exc)
@@ -145,6 +145,7 @@ class WhatsAppAdapter(BasePlatformAdapter):
        self._bridge_log: Optional[Path] = None
        self._poll_task: Optional[asyncio.Task] = None
        self._http_session: Optional["aiohttp.ClientSession"] = None
+        self._session_lock_identity: Optional[str] = None

    def _whatsapp_require_mention(self) -> bool:
        configured = self.config.extra.get("require_mention")
@@ -289,7 +290,23 @@ class WhatsAppAdapter(BasePlatformAdapter):
        
        # Acquire scoped lock to prevent duplicate sessions
        try:
-            if not self._acquire_platform_lock('whatsapp-session', str(self._session_path), 'WhatsApp session'):
+            from gateway.status import acquire_scoped_lock
+
+            self._session_lock_identity = str(self._session_path)
+            acquired, existing = acquire_scoped_lock(
+                "whatsapp-session",
+                self._session_lock_identity,
+                metadata={"platform": self.platform.value},
+            )
+            if not acquired:
+                owner_pid = existing.get("pid") if isinstance(existing, dict) else None
+                message = (
+                    "Another local Hermes gateway is already using this WhatsApp session"
+                    + (f" (PID {owner_pid})." if owner_pid else ".")
+                    + " Stop the other gateway before starting a second WhatsApp bridge."
+                )
+                logger.error("[%s] %s", self.name, message)
+                self._set_fatal_error("whatsapp_session_lock", message, retryable=False)
                return False
        except Exception as e:
            logger.warning("[%s] Could not acquire session lock (non-fatal): %s", self.name, e)
@@ -451,7 +468,12 @@ class WhatsAppAdapter(BasePlatformAdapter):
            return True
            
        except Exception as e:
-            self._release_platform_lock()
+            if self._session_lock_identity:
+                try:
+                    from gateway.status import release_scoped_lock
+                    release_scoped_lock("whatsapp-session", self._session_lock_identity)
+                except Exception:
+                    pass
            logger.error("[%s] Failed to start bridge: %s", self.name, e, exc_info=True)
            self._close_bridge_log()
            return False
@@ -524,11 +546,17 @@ class WhatsAppAdapter(BasePlatformAdapter):
            await self._http_session.close()
        self._http_session = None

-        self._release_platform_lock()
+        if self._session_lock_identity:
+            try:
+                from gateway.status import release_scoped_lock
+                release_scoped_lock("whatsapp-session", self._session_lock_identity)
+            except Exception as e:
+                logger.warning("[%s] Error releasing WhatsApp session lock: %s", self.name, e, exc_info=True)

        self._mark_disconnected()
        self._bridge_process = None
        self._close_bridge_log()
+        self._session_lock_identity = None
        print(f"[{self.name}] Disconnected")
    
    async def send(
@@ -368,11 +368,6 @@ class SessionEntry:
    # survives gateway restarts (the old in-memory _pre_flushed_sessions
    # set was lost on restart, causing redundant re-flushes).
    memory_flushed: bool = False
-
-    # When True the next call to get_or_create_session() will auto-reset
-    # this session (create a new session_id) so the user starts fresh.
-    # Set by /stop to break stuck-resume loops (#7536).
-    suspended: bool = False
    
    def to_dict(self) -> Dict[str, Any]:
        result = {
@@ -392,7 +387,6 @@ class SessionEntry:
            "estimated_cost_usd": self.estimated_cost_usd,
            "cost_status": self.cost_status,
            "memory_flushed": self.memory_flushed,
-            "suspended": self.suspended,
        }
        if self.origin:
            result["origin"] = self.origin.to_dict()
@@ -429,7 +423,6 @@ class SessionEntry:
            estimated_cost_usd=data.get("estimated_cost_usd", 0.0),
            cost_status=data.get("cost_status", "unknown"),
            memory_flushed=data.get("memory_flushed", False),
-            suspended=data.get("suspended", False),
        )


@@ -705,12 +698,7 @@ class SessionStore:
            if session_key in self._entries and not force_new:
                entry = self._entries[session_key]

-                # Auto-reset sessions marked as suspended (e.g. after /stop
-                # broke a stuck loop — #7536).
-                if entry.suspended:
-                    reset_reason = "suspended"
-                else:
-                    reset_reason = self._should_reset(entry, source)
+                reset_reason = self._should_reset(entry, source)
                if not reset_reason:
                    entry.updated_at = now
                    self._save()
@@ -783,44 +771,6 @@ class SessionStore:
                    entry.last_prompt_tokens = last_prompt_tokens
                self._save()

-    def suspend_session(self, session_key: str) -> bool:
-        """Mark a session as suspended so it auto-resets on next access.
-
-        Used by ``/stop`` to prevent stuck sessions from being resumed
-        after a gateway restart (#7536).  Returns True if the session
-        existed and was marked.
-        """
-        with self._lock:
-            self._ensure_loaded_locked()
-            if session_key in self._entries:
-                self._entries[session_key].suspended = True
-                self._save()
-                return True
-        return False
-
-    def suspend_recently_active(self, max_age_seconds: int = 120) -> int:
-        """Mark recently-active sessions as suspended.
-
-        Called on gateway startup to prevent sessions that were likely
-        in-flight when the gateway last exited from being blindly resumed
-        (#7536).  Only suspends sessions updated within *max_age_seconds*
-        to avoid resetting long-idle sessions that are harmless to resume.
-        Returns the number of sessions that were suspended.
-        """
-        from datetime import timedelta
-
-        cutoff = _now() - timedelta(seconds=max_age_seconds)
-        count = 0
-        with self._lock:
-            self._ensure_loaded_locked()
-            for entry in self._entries.values():
-                if not entry.suspended and entry.updated_at >= cutoff:
-                    entry.suspended = True
-                    count += 1
-            if count:
-                self._save()
-        return count
-
    def reset_session(self, session_key: str) -> Optional[SessionEntry]:
        """Force reset a session, creating a new session ID."""
        db_end_session_id = None
@@ -46,18 +46,12 @@ _SESSION_PLATFORM: ContextVar[str] = ContextVar("HERMES_SESSION_PLATFORM", defau
 _SESSION_CHAT_ID: ContextVar[str] = ContextVar("HERMES_SESSION_CHAT_ID", default="")
 _SESSION_CHAT_NAME: ContextVar[str] = ContextVar("HERMES_SESSION_CHAT_NAME", default="")
 _SESSION_THREAD_ID: ContextVar[str] = ContextVar("HERMES_SESSION_THREAD_ID", default="")
-_SESSION_USER_ID: ContextVar[str] = ContextVar("HERMES_SESSION_USER_ID", default="")
-_SESSION_USER_NAME: ContextVar[str] = ContextVar("HERMES_SESSION_USER_NAME", default="")
-_SESSION_KEY: ContextVar[str] = ContextVar("HERMES_SESSION_KEY", default="")

 _VAR_MAP = {
    "HERMES_SESSION_PLATFORM": _SESSION_PLATFORM,
    "HERMES_SESSION_CHAT_ID": _SESSION_CHAT_ID,
    "HERMES_SESSION_CHAT_NAME": _SESSION_CHAT_NAME,
    "HERMES_SESSION_THREAD_ID": _SESSION_THREAD_ID,
-    "HERMES_SESSION_USER_ID": _SESSION_USER_ID,
-    "HERMES_SESSION_USER_NAME": _SESSION_USER_NAME,
-    "HERMES_SESSION_KEY": _SESSION_KEY,
 }


@@ -66,9 +60,6 @@ def set_session_vars(
    chat_id: str = "",
    chat_name: str = "",
    thread_id: str = "",
-    user_id: str = "",
-    user_name: str = "",
-    session_key: str = "",
 ) -> list:
    """Set all session context variables and return reset tokens.

@@ -83,9 +74,6 @@ def set_session_vars(
        _SESSION_CHAT_ID.set(chat_id),
        _SESSION_CHAT_NAME.set(chat_name),
        _SESSION_THREAD_ID.set(thread_id),
-        _SESSION_USER_ID.set(user_id),
-        _SESSION_USER_NAME.set(user_name),
-        _SESSION_KEY.set(session_key),
    ]
    return tokens

@@ -99,9 +87,6 @@ def clear_session_vars(tokens: list) -> None:
        _SESSION_CHAT_ID,
        _SESSION_CHAT_NAME,
        _SESSION_THREAD_ID,
-        _SESSION_USER_ID,
-        _SESSION_USER_NAME,
-        _SESSION_KEY,
    ]
    for var, token in zip(vars_in_order, tokens):
        var.reset(token)
@@ -26,7 +26,6 @@ _GATEWAY_KIND = "hermes-gateway"
 _RUNTIME_STATUS_FILE = "gateway_state.json"
 _LOCKS_DIRNAME = "gateway-locks"
 _IS_WINDOWS = sys.platform == "win32"
-_UNSET = object()


 def _get_pid_path() -> Path:
@@ -219,14 +218,14 @@ def write_pid_file() -> None:

 def write_runtime_status(
    *,
-    gateway_state: Any = _UNSET,
-    exit_reason: Any = _UNSET,
-    restart_requested: Any = _UNSET,
-    active_agents: Any = _UNSET,
-    platform: Any = _UNSET,
-    platform_state: Any = _UNSET,
-    error_code: Any = _UNSET,
-    error_message: Any = _UNSET,
+    gateway_state: Optional[str] = None,
+    exit_reason: Optional[str] = None,
+    restart_requested: Optional[bool] = None,
+    active_agents: Optional[int] = None,
+    platform: Optional[str] = None,
+    platform_state: Optional[str] = None,
+    error_code: Optional[str] = None,
+    error_message: Optional[str] = None,
 ) -> None:
    """Persist gateway runtime health information for diagnostics/status."""
    path = _get_runtime_status_path()
@@ -237,22 +236,22 @@ def write_runtime_status(
    payload["start_time"] = _get_process_start_time(os.getpid())
    payload["updated_at"] = _utc_now_iso()

-    if gateway_state is not _UNSET:
+    if gateway_state is not None:
        payload["gateway_state"] = gateway_state
-    if exit_reason is not _UNSET:
+    if exit_reason is not None:
        payload["exit_reason"] = exit_reason
-    if restart_requested is not _UNSET:
+    if restart_requested is not None:
        payload["restart_requested"] = bool(restart_requested)
-    if active_agents is not _UNSET:
+    if active_agents is not None:
        payload["active_agents"] = max(0, int(active_agents))

-    if platform is not _UNSET:
+    if platform is not None:
        platform_payload = payload["platforms"].get(platform, {})
-        if platform_state is not _UNSET:
+        if platform_state is not None:
            platform_payload["state"] = platform_state
-        if error_code is not _UNSET:
+        if error_code is not None:
            platform_payload["error_code"] = error_code
-        if error_message is not _UNSET:
+        if error_message is not None:
            platform_payload["error_message"] = error_message
        platform_payload["updated_at"] = _utc_now_iso()
        payload["platforms"][platform] = platform_payload
@@ -32,15 +32,11 @@ _DONE = object()
 # new one so that subsequent text appears below tool progress messages.
 _NEW_SEGMENT = object()

-# Queue marker for a completed assistant commentary message emitted between
-# API/tool iterations (for example: "I'll inspect the repo first.").
-_COMMENTARY = object()
-

@dataclass
 class StreamConsumerConfig:
    """Runtime config for a single stream consumer instance."""
-    edit_interval: float = 1.0
+    edit_interval: float = 0.3
    buffer_threshold: int = 40
    cursor: str = " ▉"

@@ -60,10 +56,6 @@ class GatewayStreamConsumer:
        await task         # wait for final edit
    """

-    # After this many consecutive flood-control failures, permanently disable
-    # progressive edits for the remainder of the stream.
-    _MAX_FLOOD_STRIKES = 3
-
    def __init__(
        self,
        adapter: Any,
@@ -79,43 +71,18 @@ class GatewayStreamConsumer:
        self._accumulated = ""
        self._message_id: Optional[str] = None
        self._already_sent = False
-        self._edit_supported = True  # Disabled when progressive edits are no longer usable
+        self._edit_supported = True  # Disabled on first edit failure (Signal/Email/HA)
        self._last_edit_time = 0.0
        self._last_sent_text = ""   # Track last-sent text to skip redundant edits
        self._fallback_final_send = False
        self._fallback_prefix = ""
-        self._flood_strikes = 0         # Consecutive flood-control edit failures
-        self._current_edit_interval = self.cfg.edit_interval  # Adaptive backoff
-        self._final_response_sent = False

    @property
    def already_sent(self) -> bool:
-        """True if at least one message was sent or edited during the run."""
+        """True if at least one message was sent/edited — signals the base
+        adapter to skip re-sending the final response."""
        return self._already_sent

-    @property
-    def final_response_sent(self) -> bool:
-        """True when the stream consumer delivered the final assistant reply."""
-        return self._final_response_sent
-
-    def on_segment_break(self) -> None:
-        """Finalize the current stream segment and start a fresh message."""
-        self._queue.put(_NEW_SEGMENT)
-
-    def on_commentary(self, text: str) -> None:
-        """Queue a completed interim assistant commentary message."""
-        if text:
-            self._queue.put((_COMMENTARY, text))
-
-    def _reset_segment_state(self, *, preserve_no_edit: bool = False) -> None:
-        if preserve_no_edit and self._message_id == "__no_edit__":
-            return
-        self._message_id = None
-        self._accumulated = ""
-        self._last_sent_text = ""
-        self._fallback_final_send = False
-        self._fallback_prefix = ""
-
    def on_delta(self, text: str) -> None:
        """Thread-safe callback — called from the agent's worker thread.

@@ -126,7 +93,7 @@ class GatewayStreamConsumer:
        if text:
            self._queue.put(text)
        elif text is None:
-            self.on_segment_break()
+            self._queue.put(_NEW_SEGMENT)

    def finish(self) -> None:
        """Signal that the stream is complete."""
@@ -143,7 +110,6 @@ class GatewayStreamConsumer:
                # Drain all available items from the queue
                got_done = False
                got_segment_break = False
-                commentary_text = None
                while True:
                    try:
                        item = self._queue.get_nowait()
@@ -153,9 +119,6 @@ class GatewayStreamConsumer:
                        if item is _NEW_SEGMENT:
                            got_segment_break = True
                            break
-                        if isinstance(item, tuple) and len(item) == 2 and item[0] is _COMMENTARY:
-                            commentary_text = item[1]
-                            break
                        self._accumulated += item
                    except queue.Empty:
                        break
@@ -166,13 +129,11 @@ class GatewayStreamConsumer:
                should_edit = (
                    got_done
                    or got_segment_break
-                    or commentary_text is not None
-                    or (elapsed >= self._current_edit_interval
+                    or (elapsed >= self.cfg.edit_interval
                        and self._accumulated)
                    or len(self._accumulated) >= self.cfg.buffer_threshold
                )

-                current_update_visible = False
                if should_edit and self._accumulated:
                    # Split overflow: if accumulated text exceeds the platform
                    # limit, split into properly sized chunks.
@@ -194,7 +155,6 @@ class GatewayStreamConsumer:
                        self._last_sent_text = ""
                        self._last_edit_time = time.monotonic()
                        if got_done:
-                            self._final_response_sent = self._already_sent
                            return
                        if got_segment_break:
                            self._message_id = None
@@ -213,23 +173,22 @@ class GatewayStreamConsumer:
                        if split_at < _safe_limit // 2:
                            split_at = _safe_limit
                        chunk = self._accumulated[:split_at]
-                        ok = await self._send_or_edit(chunk)
-                        if self._fallback_final_send or not ok:
-                            # Edit failed (or backed off due to flood control)
-                            # while attempting to split an oversized message.
-                            # Keep the full accumulated text intact so the
-                            # fallback final-send path can deliver the remaining
-                            # continuation without dropping content.
+                        await self._send_or_edit(chunk)
+                        if self._fallback_final_send:
+                            # Edit failed while attempting to split an oversized
+                            # message. Keep the full accumulated text intact so
+                            # the fallback final-send path can deliver the
+                            # remaining continuation without dropping content.
                            break
                        self._accumulated = self._accumulated[split_at:].lstrip("\n")
                        self._message_id = None
                        self._last_sent_text = ""

                    display_text = self._accumulated
-                    if not got_done and not got_segment_break and commentary_text is None:
+                    if not got_done and not got_segment_break:
                        display_text += self.cfg.cursor

-                    current_update_visible = await self._send_or_edit(display_text)
+                    await self._send_or_edit(display_text)
                    self._last_edit_time = time.monotonic()

                if got_done:
@@ -240,20 +199,12 @@ class GatewayStreamConsumer:
                    if self._accumulated:
                        if self._fallback_final_send:
                            await self._send_fallback_final(self._accumulated)
-                        elif current_update_visible:
-                            self._final_response_sent = True
                        elif self._message_id:
-                            self._final_response_sent = await self._send_or_edit(self._accumulated)
+                            await self._send_or_edit(self._accumulated)
                        elif not self._already_sent:
-                            self._final_response_sent = await self._send_or_edit(self._accumulated)
+                            await self._send_or_edit(self._accumulated)
                    return

-                if commentary_text is not None:
-                    self._reset_segment_state()
-                    await self._send_commentary(commentary_text)
-                    self._last_edit_time = time.monotonic()
-                    self._reset_segment_state()
-
                # Tool boundary: reset message state so the next text chunk
                # creates a fresh message below any tool-progress messages.
                #
@@ -262,14 +213,17 @@ class GatewayStreamConsumer:
                # github_comment delivery).  Resetting to None would re-enter
                # the "first send" path on every tool boundary and post one
                # platform message per tool call — that is what caused 155
-                # comments under a single PR.  Instead, preserve the sentinel
-                # so the full continuation is delivered once via
-                # _send_fallback_final.
+                # comments under a single PR.  Instead, keep all state so the
+                # full continuation is delivered once via _send_fallback_final.
                # (When editing fails mid-stream due to flood control the id is
                # a real string like "msg_1", not "__no_edit__", so that case
                # still resets and creates a fresh segment as intended.)
-                if got_segment_break:
-                    self._reset_segment_state(preserve_no_edit=True)
+                if got_segment_break and self._message_id != "__no_edit__":
+                    self._message_id = None
+                    self._accumulated = ""
+                    self._last_sent_text = ""
+                    self._fallback_final_send = False
+                    self._fallback_prefix = ""

                await asyncio.sleep(0.05)  # Small yield to not busy-loop

@@ -368,17 +322,13 @@ class GatewayStreamConsumer:
        return chunks

    async def _send_fallback_final(self, text: str) -> None:
-        """Send the final continuation after streaming edits stop working.
-
-        Retries each chunk once on flood-control failures with a short delay.
-        """
+        """Send the final continuation after streaming edits stop working."""
        final_text = self._clean_for_display(text)
        continuation = self._continuation_text(final_text)
        self._fallback_final_send = False
        if not continuation.strip():
            # Nothing new to send — the visible partial already matches final text.
            self._already_sent = True
-            self._final_response_sent = True
            return

        raw_limit = getattr(self.adapter, "MAX_MESSAGE_LENGTH", 4096)
@@ -389,31 +339,17 @@ class GatewayStreamConsumer:
        last_successful_chunk = ""
        sent_any_chunk = False
        for chunk in chunks:
-            # Try sending with one retry on flood-control errors.
-            result = None
-            for attempt in range(2):
-                result = await self.adapter.send(
-                    chat_id=self.chat_id,
-                    content=chunk,
-                    metadata=self.metadata,
-                )
-                if result.success:
-                    break
-                if attempt == 0 and self._is_flood_error(result):
-                    logger.debug(
-                        "Flood control on fallback send, retrying in 3s"
-                    )
-                    await asyncio.sleep(3.0)
-                else:
-                    break  # non-flood error or second attempt failed
-
-            if not result or not result.success:
+            result = await self.adapter.send(
+                chat_id=self.chat_id,
+                content=chunk,
+                metadata=self.metadata,
+            )
+            if not result.success:
                if sent_any_chunk:
                    # Some continuation text already reached the user. Suppress
                    # the base gateway final-send path so we don't resend the
                    # full response and create another duplicate.
                    self._already_sent = True
-                    self._final_response_sent = True
                    self._message_id = last_message_id
                    self._last_sent_text = last_successful_chunk
                    self._fallback_prefix = ""
@@ -431,74 +367,23 @@ class GatewayStreamConsumer:

        self._message_id = last_message_id
        self._already_sent = True
-        self._final_response_sent = True
        self._last_sent_text = chunks[-1]
        self._fallback_prefix = ""

-    def _is_flood_error(self, result) -> bool:
-        """Check if a SendResult failure is due to flood control / rate limiting."""
-        err = getattr(result, "error", "") or ""
-        err_lower = err.lower()
-        return "flood" in err_lower or "retry after" in err_lower or "rate" in err_lower
-
-    async def _try_strip_cursor(self) -> None:
-        """Best-effort edit to remove the cursor from the last visible message.
-
-        Called when entering fallback mode so the user doesn't see a stuck
-        cursor (▉) in the partial message.
-        """
-        if not self._message_id or self._message_id == "__no_edit__":
-            return
-        prefix = self._visible_prefix()
-        if not prefix or not prefix.strip():
-            return
-        try:
-            await self.adapter.edit_message(
-                chat_id=self.chat_id,
-                message_id=self._message_id,
-                content=prefix,
-            )
-            self._last_sent_text = prefix
-        except Exception:
-            pass  # best-effort — don't let this block the fallback path
-
-    async def _send_commentary(self, text: str) -> bool:
-        """Send a completed interim assistant commentary message."""
-        text = self._clean_for_display(text)
-        if not text.strip():
-            return False
-        try:
-            result = await self.adapter.send(
-                chat_id=self.chat_id,
-                content=text,
-                metadata=self.metadata,
-            )
-            if result.success:
-                self._already_sent = True
-                return True
-        except Exception as e:
-            logger.error("Commentary send error: %s", e)
-        return False
-
-    async def _send_or_edit(self, text: str) -> bool:
-        """Send or edit the streaming message.
-
-        Returns True if the text was successfully delivered (sent or edited),
-        False otherwise.  Callers like the overflow split loop use this to
-        decide whether to advance past the delivered chunk.
-        """
+    async def _send_or_edit(self, text: str) -> None:
+        """Send or edit the streaming message."""
        # Strip MEDIA: directives so they don't appear as visible text.
        # Media files are delivered as native attachments after the stream
        # finishes (via _deliver_media_from_response in gateway/run.py).
        text = self._clean_for_display(text)
        if not text.strip():
-            return True  # nothing to send is "success"
+            return
        try:
            if self._message_id is not None:
                if self._edit_supported:
                    # Skip if text is identical to what we last sent
                    if text == self._last_sent_text:
-                        return True
+                        return
                    # Edit existing message
                    result = await self.adapter.edit_message(
                        chat_id=self.chat_id,
@@ -508,52 +393,19 @@ class GatewayStreamConsumer:
                    if result.success:
                        self._already_sent = True
                        self._last_sent_text = text
-                        # Successful edit — reset flood strike counter
-                        self._flood_strikes = 0
-                        return True
                    else:
-                        # Edit failed.  If this looks like flood control / rate
-                        # limiting, use adaptive backoff: double the edit interval
-                        # and retry on the next cycle.  Only permanently disable
-                        # edits after _MAX_FLOOD_STRIKES consecutive failures.
-                        if self._is_flood_error(result):
-                            self._flood_strikes += 1
-                            self._current_edit_interval = min(
-                                self._current_edit_interval * 2, 10.0,
-                            )
-                            logger.debug(
-                                "Flood control on edit (strike %d/%d), "
-                                "backoff interval → %.1fs",
-                                self._flood_strikes,
-                                self._MAX_FLOOD_STRIKES,
-                                self._current_edit_interval,
-                            )
-                            if self._flood_strikes < self._MAX_FLOOD_STRIKES:
-                                # Don't disable edits yet — just slow down.
-                                # Update _last_edit_time so the next edit
-                                # respects the new interval.
-                                self._last_edit_time = time.monotonic()
-                                return False
-
-                        # Non-flood error OR flood strikes exhausted: enter
-                        # fallback mode — send only the missing tail once the
+                        # If an edit fails mid-stream (especially Telegram flood control),
+                        # stop progressive edits and send only the missing tail once the
                        # final response is available.
-                        logger.debug(
-                            "Edit failed (strikes=%d), entering fallback mode",
-                            self._flood_strikes,
-                        )
+                        logger.debug("Edit failed, disabling streaming for this adapter")
                        self._fallback_prefix = self._visible_prefix()
                        self._fallback_final_send = True
                        self._edit_supported = False
                        self._already_sent = True
-                        # Best-effort: strip the cursor from the last visible
-                        # message so the user doesn't see a stuck ▉.
-                        await self._try_strip_cursor()
-                        return False
                else:
                    # Editing not supported — skip intermediate updates.
                    # The final response will be sent by the fallback path.
-                    return False
+                    pass
            else:
                # First message — send new
                result = await self.adapter.send(
@@ -561,25 +413,23 @@ class GatewayStreamConsumer:
                    content=text,
                    metadata=self.metadata,
                )
-                if result.success:
-                    if result.message_id:
-                        self._message_id = result.message_id
-                    else:
-                        self._edit_supported = False
+                if result.success and result.message_id:
+                    self._message_id = result.message_id
                    self._already_sent = True
                    self._last_sent_text = text
-                    if not result.message_id:
-                        self._fallback_prefix = self._visible_prefix()
-                        self._fallback_final_send = True
-                        # Sentinel prevents re-entering the first-send path on
-                        # every delta/tool boundary when platforms accept a
-                        # message but do not return an editable message id.
-                        self._message_id = "__no_edit__"
-                    return True
+                elif result.success:
+                    # Platform accepted the message but returned no message_id
+                    # (e.g. Signal).  Can't edit without an ID — switch to
+                    # fallback mode: suppress intermediate deltas, send only
+                    # the missing tail once the final response is ready.
+                    self._already_sent = True
+                    self._edit_supported = False
+                    self._fallback_prefix = self._clean_for_display(text)
+                    self._fallback_final_send = True
+                    # Sentinel prevents re-entering this branch on every delta
+                    self._message_id = "__no_edit__"
                else:
                    # Initial send failed — disable streaming for this session
                    self._edit_supported = False
-                    return False
        except Exception as e:
            logger.error("Stream send/edit error: %s", e)
-            return False
@@ -250,39 +250,9 @@ PROVIDER_REGISTRY: Dict[str, ProviderConfig] = {
        api_key_env_vars=("HF_TOKEN",),
        base_url_env_var="HF_BASE_URL",
    ),
-    "xiaomi": ProviderConfig(
-        id="xiaomi",
-        name="Xiaomi MiMo",
-        auth_type="api_key",
-        inference_base_url="https://api.xiaomimimo.com/v1",
-        api_key_env_vars=("XIAOMI_API_KEY",),
-        base_url_env_var="XIAOMI_BASE_URL",
-    ),
 }


-# =============================================================================
-# Anthropic Key Helper
-# =============================================================================
-
-def get_anthropic_key() -> str:
-    """Return the first usable Anthropic credential, or ``""``.
-
-    Checks both the ``.env`` file (via ``get_env_value``) and the process
-    environment (``os.getenv``).  The fallback order mirrors the
-    ``PROVIDER_REGISTRY["anthropic"].api_key_env_vars`` tuple:
-
-        ANTHROPIC_API_KEY -> ANTHROPIC_TOKEN -> CLAUDE_CODE_OAUTH_TOKEN
-    """
-    from hermes_cli.config import get_env_value
-
-    for var in PROVIDER_REGISTRY["anthropic"].api_key_env_vars:
-        value = get_env_value(var) or os.getenv(var, "")
-        if value:
-            return value
-    return ""
-
-
 # =============================================================================
 # Kimi Code Endpoint Detection
 # =============================================================================
@@ -938,7 +908,6 @@ def resolve_provider(
        "opencode": "opencode-zen", "zen": "opencode-zen",
        "qwen-portal": "qwen-oauth", "qwen-cli": "qwen-oauth", "qwen-oauth": "qwen-oauth",
        "hf": "huggingface", "hugging-face": "huggingface", "huggingface-hub": "huggingface",
-        "mimo": "xiaomi", "xiaomi-mimo": "xiaomi",
        "go": "opencode-go", "opencode-go-sub": "opencode-go",
        "kilo": "kilocode", "kilo-code": "kilocode", "kilo-gateway": "kilocode",
        # Local server aliases — route through the generic custom provider
@@ -1,399 +0,0 @@
-"""
-Backup and import commands for hermes CLI.
-
-`hermes backup` creates a zip archive of the entire ~/.hermes/ directory
-(excluding the hermes-agent repo and transient files).
-
-`hermes import` restores from a backup zip, overlaying onto the current
-HERMES_HOME root.
-"""
-
-import os
-import sys
-import time
-import zipfile
-from datetime import datetime
-from pathlib import Path
-
-from hermes_constants import get_default_hermes_root, display_hermes_home
-
-
-# ---------------------------------------------------------------------------
-# Exclusion rules
-# ---------------------------------------------------------------------------
-
-# Directory names to skip entirely (matched against each path component)
-_EXCLUDED_DIRS = {
-    "hermes-agent",     # the codebase repo — re-clone instead
-    "__pycache__",      # bytecode caches — regenerated on import
-    ".git",             # nested git dirs (profiles shouldn't have these, but safety)
-    "node_modules",     # js deps if website/ somehow leaks in
-}
-
-# File-name suffixes to skip
-_EXCLUDED_SUFFIXES = (
-    ".pyc",
-    ".pyo",
-)
-
-# File names to skip (runtime state that's meaningless on another machine)
-_EXCLUDED_NAMES = {
-    "gateway.pid",
-    "cron.pid",
-}
-
-
-def _should_exclude(rel_path: Path) -> bool:
-    """Return True if *rel_path* (relative to hermes root) should be skipped."""
-    parts = rel_path.parts
-
-    # Any path component matches an excluded dir name
-    for part in parts:
-        if part in _EXCLUDED_DIRS:
-            return True
-
-    name = rel_path.name
-
-    if name in _EXCLUDED_NAMES:
-        return True
-
-    if name.endswith(_EXCLUDED_SUFFIXES):
-        return True
-
-    return False
-
-
-# ---------------------------------------------------------------------------
-# Backup
-# ---------------------------------------------------------------------------
-
-def _format_size(nbytes: int) -> str:
-    """Human-readable file size."""
-    for unit in ("B", "KB", "MB", "GB"):
-        if nbytes < 1024:
-            return f"{nbytes:.1f} {unit}" if unit != "B" else f"{nbytes} {unit}"
-        nbytes /= 1024
-    return f"{nbytes:.1f} TB"
-
-
-def run_backup(args) -> None:
-    """Create a zip backup of the Hermes home directory."""
-    hermes_root = get_default_hermes_root()
-
-    if not hermes_root.is_dir():
-        print(f"Error: Hermes home directory not found at {hermes_root}")
-        sys.exit(1)
-
-    # Determine output path
-    if args.output:
-        out_path = Path(args.output).expanduser().resolve()
-        # If user gave a directory, put the zip inside it
-        if out_path.is_dir():
-            stamp = datetime.now().strftime("%Y-%m-%d-%H%M%S")
-            out_path = out_path / f"hermes-backup-{stamp}.zip"
-    else:
-        stamp = datetime.now().strftime("%Y-%m-%d-%H%M%S")
-        out_path = Path.home() / f"hermes-backup-{stamp}.zip"
-
-    # Ensure the suffix is .zip
-    if out_path.suffix.lower() != ".zip":
-        out_path = out_path.with_suffix(out_path.suffix + ".zip")
-
-    # Ensure parent directory exists
-    out_path.parent.mkdir(parents=True, exist_ok=True)
-
-    # Collect files
-    print(f"Scanning {display_hermes_home()} ...")
-    files_to_add: list[tuple[Path, Path]] = []  # (absolute, relative)
-    skipped_dirs = set()
-
-    for dirpath, dirnames, filenames in os.walk(hermes_root, followlinks=False):
-        dp = Path(dirpath)
-        rel_dir = dp.relative_to(hermes_root)
-
-        # Prune excluded directories in-place so os.walk doesn't descend
-        orig_dirnames = dirnames[:]
-        dirnames[:] = [
-            d for d in dirnames
-            if d not in _EXCLUDED_DIRS
-        ]
-        for removed in set(orig_dirnames) - set(dirnames):
-            skipped_dirs.add(str(rel_dir / removed))
-
-        for fname in filenames:
-            fpath = dp / fname
-            rel = fpath.relative_to(hermes_root)
-
-            if _should_exclude(rel):
-                continue
-
-            # Skip the output zip itself if it happens to be inside hermes root
-            try:
-                if fpath.resolve() == out_path.resolve():
-                    continue
-            except (OSError, ValueError):
-                pass
-
-            files_to_add.append((fpath, rel))
-
-    if not files_to_add:
-        print("No files to back up.")
-        return
-
-    # Create the zip
-    file_count = len(files_to_add)
-    print(f"Backing up {file_count} files ...")
-
-    total_bytes = 0
-    errors = []
-    t0 = time.monotonic()
-
-    with zipfile.ZipFile(out_path, "w", zipfile.ZIP_DEFLATED, compresslevel=6) as zf:
-        for i, (abs_path, rel_path) in enumerate(files_to_add, 1):
-            try:
-                zf.write(abs_path, arcname=str(rel_path))
-                total_bytes += abs_path.stat().st_size
-            except (PermissionError, OSError) as exc:
-                errors.append(f"  {rel_path}: {exc}")
-                continue
-
-            # Progress every 500 files
-            if i % 500 == 0:
-                print(f"  {i}/{file_count} files ...")
-
-    elapsed = time.monotonic() - t0
-    zip_size = out_path.stat().st_size
-
-    # Summary
-    print()
-    print(f"Backup complete: {out_path}")
-    print(f"  Files:       {file_count}")
-    print(f"  Original:    {_format_size(total_bytes)}")
-    print(f"  Compressed:  {_format_size(zip_size)}")
-    print(f"  Time:        {elapsed:.1f}s")
-
-    if skipped_dirs:
-        print(f"\n  Excluded directories:")
-        for d in sorted(skipped_dirs):
-            print(f"    {d}/")
-
-    if errors:
-        print(f"\n  Warnings ({len(errors)} files skipped):")
-        for e in errors[:10]:
-            print(e)
-        if len(errors) > 10:
-            print(f"  ... and {len(errors) - 10} more")
-
-    print(f"\nRestore with: hermes import {out_path.name}")
-
-
-# ---------------------------------------------------------------------------
-# Import
-# ---------------------------------------------------------------------------
-
-def _validate_backup_zip(zf: zipfile.ZipFile) -> tuple[bool, str]:
-    """Check that a zip looks like a Hermes backup.
-
-    Returns (ok, reason).
-    """
-    names = zf.namelist()
-    if not names:
-        return False, "zip archive is empty"
-
-    # Look for telltale files that a hermes home would have
-    markers = {"config.yaml", ".env", "hermes_state.db", "memory_store.db"}
-    found = set()
-    for n in names:
-        # Could be at the root or one level deep (if someone zipped the directory)
-        basename = Path(n).name
-        if basename in markers:
-            found.add(basename)
-
-    if not found:
-        return False, (
-            "zip does not appear to be a Hermes backup "
-            "(no config.yaml, .env, or state databases found)"
-        )
-
-    return True, ""
-
-
-def _detect_prefix(zf: zipfile.ZipFile) -> str:
-    """Detect if the zip has a common directory prefix wrapping all entries.
-
-    Some tools zip as `.hermes/config.yaml` instead of `config.yaml`.
-    Returns the prefix to strip (empty string if none).
-    """
-    names = [n for n in zf.namelist() if not n.endswith("/")]
-    if not names:
-        return ""
-
-    # Find common prefix
-    parts_list = [Path(n).parts for n in names]
-
-    # Check if all entries share a common first directory
-    first_parts = {p[0] for p in parts_list if len(p) > 1}
-    if len(first_parts) == 1:
-        prefix = first_parts.pop()
-        # Only strip if it looks like a hermes dir name
-        if prefix in (".hermes", "hermes"):
-            return prefix + "/"
-
-    return ""
-
-
-def run_import(args) -> None:
-    """Restore a Hermes backup from a zip file."""
-    zip_path = Path(args.zipfile).expanduser().resolve()
-
-    if not zip_path.is_file():
-        print(f"Error: File not found: {zip_path}")
-        sys.exit(1)
-
-    if not zipfile.is_zipfile(zip_path):
-        print(f"Error: Not a valid zip file: {zip_path}")
-        sys.exit(1)
-
-    hermes_root = get_default_hermes_root()
-
-    with zipfile.ZipFile(zip_path, "r") as zf:
-        # Validate
-        ok, reason = _validate_backup_zip(zf)
-        if not ok:
-            print(f"Error: {reason}")
-            sys.exit(1)
-
-        prefix = _detect_prefix(zf)
-        members = [n for n in zf.namelist() if not n.endswith("/")]
-        file_count = len(members)
-
-        print(f"Backup contains {file_count} files")
-        print(f"Target: {display_hermes_home()}")
-
-        if prefix:
-            print(f"Detected archive prefix: {prefix!r} (will be stripped)")
-
-        # Check for existing installation
-        has_config = (hermes_root / "config.yaml").exists()
-        has_env = (hermes_root / ".env").exists()
-
-        if (has_config or has_env) and not args.force:
-            print()
-            print("Warning: Target directory already has Hermes configuration.")
-            print("Importing will overwrite existing files with backup contents.")
-            print()
-            try:
-                answer = input("Continue? [y/N] ").strip().lower()
-            except (EOFError, KeyboardInterrupt):
-                print("\nAborted.")
-                sys.exit(1)
-            if answer not in ("y", "yes"):
-                print("Aborted.")
-                return
-
-        # Extract
-        print(f"\nImporting {file_count} files ...")
-        hermes_root.mkdir(parents=True, exist_ok=True)
-
-        errors = []
-        restored = 0
-        t0 = time.monotonic()
-
-        for member in members:
-            # Strip prefix if detected
-            if prefix and member.startswith(prefix):
-                rel = member[len(prefix):]
-            else:
-                rel = member
-
-            if not rel:
-                continue
-
-            target = hermes_root / rel
-
-            # Security: reject absolute paths and traversals
-            try:
-                target.resolve().relative_to(hermes_root.resolve())
-            except ValueError:
-                errors.append(f"  {rel}: path traversal blocked")
-                continue
-
-            try:
-                target.parent.mkdir(parents=True, exist_ok=True)
-                with zf.open(member) as src, open(target, "wb") as dst:
-                    dst.write(src.read())
-                restored += 1
-            except (PermissionError, OSError) as exc:
-                errors.append(f"  {rel}: {exc}")
-
-            if restored % 500 == 0:
-                print(f"  {restored}/{file_count} files ...")
-
-        elapsed = time.monotonic() - t0
-
-        # Summary
-        print()
-        print(f"Import complete: {restored} files restored in {elapsed:.1f}s")
-        print(f"  Target: {display_hermes_home()}")
-
-        if errors:
-            print(f"\n  Warnings ({len(errors)} files skipped):")
-            for e in errors[:10]:
-                print(e)
-            if len(errors) > 10:
-                print(f"  ... and {len(errors) - 10} more")
-
-        # Post-import: restore profile wrapper scripts
-        profiles_dir = hermes_root / "profiles"
-        restored_profiles = []
-        if profiles_dir.is_dir():
-            try:
-                from hermes_cli.profiles import (
-                    create_wrapper_script, check_alias_collision,
-                    _is_wrapper_dir_in_path, _get_wrapper_dir,
-                )
-                for entry in sorted(profiles_dir.iterdir()):
-                    if not entry.is_dir():
-                        continue
-                    profile_name = entry.name
-                    # Only create wrappers for directories with config
-                    if not (entry / "config.yaml").exists() and not (entry / ".env").exists():
-                        continue
-                    collision = check_alias_collision(profile_name)
-                    if collision:
-                        print(f"  Skipped alias '{profile_name}': {collision}")
-                        restored_profiles.append((profile_name, False))
-                    else:
-                        wrapper = create_wrapper_script(profile_name)
-                        restored_profiles.append((profile_name, wrapper is not None))
-
-                if restored_profiles:
-                    created = [n for n, ok in restored_profiles if ok]
-                    skipped = [n for n, ok in restored_profiles if not ok]
-                    if created:
-                        print(f"\n  Profile aliases restored: {', '.join(created)}")
-                    if skipped:
-                        print(f"  Profile aliases skipped:  {', '.join(skipped)}")
-                    if not _is_wrapper_dir_in_path():
-                        print(f"\n  Note: {_get_wrapper_dir()} is not in your PATH.")
-                        print('  Add to your shell config (~/.bashrc or ~/.zshrc):')
-                        print('    export PATH="$HOME/.local/bin:$PATH"')
-            except ImportError:
-                # hermes_cli.profiles might not be available (fresh install)
-                if any(profiles_dir.iterdir()):
-                    print(f"\n  Profiles detected but aliases could not be created.")
-                    print(f"  Run: hermes profile list  (after installing hermes)")
-
-        # Guidance
-        print()
-        if not (hermes_root / "hermes-agent").is_dir():
-            print("Note: The hermes-agent codebase was not included in the backup.")
-            print("  If this is a fresh install, run: hermes update")
-
-        if restored_profiles:
-            gw_profiles = [n for n, _ in restored_profiles]
-            print("\nTo re-enable gateway services for profiles:")
-            for pname in gw_profiles:
-                print(f"  hermes -p {pname} gateway install")
-
-        print("Done. Your Hermes configuration has been restored.")
@@ -1,9 +1,8 @@
 """hermes claw — OpenClaw migration commands.

 Usage:
-    hermes claw migrate              # Preview then migrate (always shows preview first)
-    hermes claw migrate --dry-run    # Preview only, no changes
-    hermes claw migrate --yes        # Skip confirmation prompt
+    hermes claw migrate              # Interactive migration from ~/.openclaw
+    hermes claw migrate --dry-run    # Preview what would be migrated
    hermes claw migrate --preset full --overwrite  # Full migration, overwrite conflicts
    hermes claw cleanup              # Archive leftover OpenClaw directories
    hermes claw cleanup --dry-run    # Preview what would be archived
@@ -50,45 +49,10 @@ _OPENCLAW_SCRIPT_INSTALLED = (
 )

 # Known OpenClaw directory names (current + legacy)
-_OPENCLAW_DIR_NAMES = (".openclaw", ".clawdbot", ".moltbot")
+_OPENCLAW_DIR_NAMES = (".openclaw", ".clawdbot", ".moldbot")

-def _warn_if_gateway_running(auto_yes: bool) -> None:
-    """Check if a Hermes gateway is running with connected platforms.
-
-    Migrating bot tokens while the gateway is polling will cause conflicts
-    (e.g. Telegram 409 "terminated by other getUpdates request"). Warn the
-    user and let them decide whether to continue.
-    """
-    from gateway.status import get_running_pid, read_runtime_status
-
-    if not get_running_pid():
-        return
-
-    data = read_runtime_status() or {}
-    platforms = data.get("platforms") or {}
-    connected = [name for name, info in platforms.items()
-                 if isinstance(info, dict) and info.get("state") == "connected"]
-    if not connected:
-        return
-
-    print()
-    print_error(
-        "Hermes gateway is running with active connections: "
-        + ", ".join(connected)
-    )
-    print_info(
-        "Migrating bot tokens while the gateway is active will cause "
-        "conflicts (Telegram, Discord, and Slack only allow one active "
-        "session per token)."
-    )
-    print_info("Recommendation: stop the gateway first with 'hermes stop'.")
-    print()
-    if not auto_yes and not prompt_yes_no("Continue anyway?", default=False):
-        print_info("Migration cancelled. Stop the gateway and try again.")
-        sys.exit(0)
-
-# State files commonly found in OpenClaw workspace directories — listed
-# during cleanup to help the user decide whether to archive
+# State files commonly found in OpenClaw workspace directories that cause
+# confusion after migration (the agent discovers them and writes to them)
 _WORKSPACE_STATE_GLOBS = (
    "*/todo.json",
    "*/sessions/*",
@@ -133,7 +97,7 @@ def _find_openclaw_dirs() -> list[Path]:


 def _scan_workspace_state(source_dir: Path) -> list[tuple[Path, str]]:
-    """Scan an OpenClaw directory for workspace state files.
+    """Scan an OpenClaw directory for workspace state files that cause confusion.

    Returns a list of (path, description) tuples.
    """
@@ -216,7 +180,7 @@ def _cmd_migrate(args):
        source_dir = Path.home() / ".openclaw"
        if not source_dir.is_dir():
            # Try legacy directory names
-            for legacy in (".clawdbot", ".moltbot"):
+            for legacy in (".clawdbot", ".moldbot"):
                candidate = Path.home() / legacy
                if candidate.is_dir():
                    source_dir = candidate
@@ -273,12 +237,12 @@ def _cmd_migrate(args):

    # Show what we're doing
    hermes_home = get_hermes_home()
-    auto_yes = getattr(args, "yes", False)
    print()
    print_header("Migration Settings")
    print_info(f"Source:      {source_dir}")
    print_info(f"Target:      {hermes_home}")
    print_info(f"Preset:      {preset}")
+    print_info(f"Mode:        {'dry run (preview only)' if dry_run else 'execute'}")
    print_info(f"Overwrite:   {'yes' if overwrite else 'no (skip conflicts)'}")
    print_info(f"Secrets:     {'yes (allowlisted only)' if migrate_secrets else 'no'}")
    if skill_conflict != "skip":
@@ -287,85 +251,31 @@ def _cmd_migrate(args):
        print_info(f"Workspace:   {workspace_target}")
    print()

-    # Check if a gateway is running with connected platforms — migrating tokens
-    # while the gateway is active will cause conflicts (e.g. Telegram 409).
-    _warn_if_gateway_running(auto_yes)
+    # For execute mode (non-dry-run), confirm unless --yes was passed
+    if not dry_run and not getattr(args, "yes", False):
+        if not prompt_yes_no("Proceed with migration?", default=True):
+            print_info("Migration cancelled.")
+            return

    # Ensure config.yaml exists before migration tries to read it
    config_path = get_config_path()
    if not config_path.exists():
        save_config(load_config())

-    # Load the migration module
+    # Load and run the migration
    try:
        mod = _load_migration_module(script_path)
        if mod is None:
            print_error("Could not load migration script.")
            return
-    except Exception as e:
-        print()
-        print_error(f"Could not load migration script: {e}")
-        logger.debug("OpenClaw migration error", exc_info=True)
-        return

-    selected = mod.resolve_selected_options(None, None, preset=preset)
-    ws_target = Path(workspace_target).resolve() if workspace_target else None
+        selected = mod.resolve_selected_options(None, None, preset=preset)
+        ws_target = Path(workspace_target).resolve() if workspace_target else None

-    # ── Phase 1: Always preview first ──────────────────────────
-    try:
-        preview = mod.Migrator(
-            source_root=source_dir.resolve(),
-            target_root=hermes_home.resolve(),
-            execute=False,
-            workspace_target=ws_target,
-            overwrite=overwrite,
-            migrate_secrets=migrate_secrets,
-            output_dir=None,
-            selected_options=selected,
-            preset_name=preset,
-            skill_conflict_mode=skill_conflict,
-        )
-        preview_report = preview.migrate()
-    except Exception as e:
-        print()
-        print_error(f"Migration preview failed: {e}")
-        logger.debug("OpenClaw migration preview error", exc_info=True)
-        return
-
-    preview_summary = preview_report.get("summary", {})
-    preview_count = preview_summary.get("migrated", 0)
-
-    if preview_count == 0:
-        print()
-        print_info("Nothing to migrate from OpenClaw.")
-        _print_migration_report(preview_report, dry_run=True)
-        return
-
-    print()
-    print_header(f"Migration Preview — {preview_count} item(s) would be imported")
-    print_info("No changes have been made yet. Review the list below:")
-    _print_migration_report(preview_report, dry_run=True)
-
-    # If --dry-run, stop here
-    if dry_run:
-        return
-
-    # ── Phase 2: Confirm and execute ───────────────────────────
-    print()
-    if not auto_yes:
-        if not sys.stdin.isatty():
-            print_info("Non-interactive session — preview only.")
-            print_info("To execute, re-run with: hermes claw migrate --yes")
-            return
-        if not prompt_yes_no("Proceed with migration?", default=True):
-            print_info("Migration cancelled.")
-            return
-
-    try:
        migrator = mod.Migrator(
            source_root=source_dir.resolve(),
            target_root=hermes_home.resolve(),
-            execute=True,
+            execute=not dry_run,
            workspace_target=ws_target,
            overwrite=overwrite,
            migrate_secrets=migrate_secrets,
@@ -382,18 +292,62 @@ def _cmd_migrate(args):
        return

    # Print results
-    _print_migration_report(report, dry_run=False)
+    _print_migration_report(report, dry_run)

-    # Source directory is left untouched — archiving is not the migration
-    # tool's responsibility.  Users who want to clean up can run
-    # 'hermes claw cleanup' separately.
+    # After successful non-dry-run migration, offer to archive the source directory
+    if not dry_run and report.get("summary", {}).get("migrated", 0) > 0:
+        _offer_source_archival(source_dir, getattr(args, "yes", False))
+
+
+def _offer_source_archival(source_dir: Path, auto_yes: bool = False):
+    """After migration, offer to rename the source directory to prevent state fragmentation.
+
+    OpenClaw workspace directories contain state files (todo.json, sessions, etc.)
+    that the agent may discover and write to, causing confusion. Renaming the
+    directory prevents this.
+    """
+    if not source_dir.is_dir():
+        return
+
+    # Scan for state files that could cause problems
+    state_files = _scan_workspace_state(source_dir)
+
+    print()
+    print_header("Post-Migration Cleanup")
+    print_info("The OpenClaw directory still exists and contains workspace state files")
+    print_info("that can confuse the agent (todo lists, sessions, logs).")
+    if state_files:
+        print()
+        print(color("  Found state files:", Colors.YELLOW))
+        # Show up to 10 most relevant findings
+        for path, desc in state_files[:10]:
+            print(f"      {desc}")
+        if len(state_files) > 10:
+            print(f"      ... and {len(state_files) - 10} more")
+    print()
+    print_info(f"Recommend: rename {source_dir.name}/ to {source_dir.name}.pre-migration/")
+    print_info("This prevents the agent from discovering old workspace directories.")
+    print_info("You can always rename it back if needed.")
+    print()
+
+    if auto_yes or prompt_yes_no(f"Archive {source_dir} now?", default=True):
+        try:
+            archive_path = _archive_directory(source_dir)
+            print_success(f"Archived: {source_dir} → {archive_path}")
+            print_info("The original directory has been renamed, not deleted.")
+            print_info(f"To undo: mv {archive_path} {source_dir}")
+        except OSError as e:
+            print_error(f"Could not archive: {e}")
+            print_info(f"You can do it manually: mv {source_dir} {source_dir}.pre-migration")
+    else:
+        print_info("Skipped. You can archive later with: hermes claw cleanup")


 def _cmd_cleanup(args):
    """Archive leftover OpenClaw directories after migration.

    Scans for OpenClaw directories that still exist after migration and offers
-    to rename them to .pre-migration to free disk space.
+    to rename them to .pre-migration to prevent state fragmentation.
    """
    dry_run = getattr(args, "dry_run", False)
    auto_yes = getattr(args, "yes", False)
@@ -468,7 +422,7 @@ def _cmd_cleanup(args):

        if state_files:
            print()
-            print(color(f"  {len(state_files)} state file(s) found:", Colors.YELLOW))
+            print(color(f"  {len(state_files)} state file(s) that could cause confusion:", Colors.YELLOW))
            for path, desc in state_files[:8]:
                print(f"      {desc}")
            if len(state_files) > 8:
@@ -479,9 +433,6 @@ def _cmd_cleanup(args):
        if dry_run:
            archive_path = _archive_directory(source_dir, dry_run=True)
            print_info(f"Would archive: {source_dir} → {archive_path}")
-        elif not auto_yes and not sys.stdin.isatty():
-            print_info(f"Non-interactive session — would archive: {source_dir}")
-            print_info("To execute, re-run with: hermes claw cleanup --yes")
        else:
            if auto_yes or prompt_yes_no(f"Archive {source_dir}?", default=True):
                try:
@@ -1,79 +0,0 @@
-"""Shared CLI output helpers for Hermes CLI modules.
-
-Extracts the identical ``print_info/success/warning/error`` and ``prompt()``
-functions previously duplicated across setup.py, tools_config.py,
-mcp_config.py, and memory_setup.py.
-"""
-
-import getpass
-import sys
-
-from hermes_cli.colors import Colors, color
-
-
-# ─── Print Helpers ────────────────────────────────────────────────────────────
-
-
-def print_info(text: str) -> None:
-    """Print a dim informational message."""
-    print(color(f"  {text}", Colors.DIM))
-
-
-def print_success(text: str) -> None:
-    """Print a green success message with ✓ prefix."""
-    print(color(f"✓ {text}", Colors.GREEN))
-
-
-def print_warning(text: str) -> None:
-    """Print a yellow warning message with ⚠ prefix."""
-    print(color(f"⚠ {text}", Colors.YELLOW))
-
-
-def print_error(text: str) -> None:
-    """Print a red error message with ✗ prefix."""
-    print(color(f"✗ {text}", Colors.RED))
-
-
-def print_header(text: str) -> None:
-    """Print a bold yellow header."""
-    print(color(f"\n  {text}", Colors.YELLOW))
-
-
-# ─── Input Prompts ────────────────────────────────────────────────────────────
-
-
-def prompt(
-    question: str,
-    default: str | None = None,
-    password: bool = False,
-) -> str:
-    """Prompt the user for input with optional default and password masking.
-
-    Replaces the four independent ``_prompt()`` / ``prompt()`` implementations
-    in setup.py, tools_config.py, mcp_config.py, and memory_setup.py.
-
-    Returns the user's input (stripped), or *default* if the user presses Enter.
-    Returns empty string on Ctrl-C or EOF.
-    """
-    suffix = f" [{default}]" if default else ""
-    display = color(f"  {question}{suffix}: ", Colors.YELLOW)
-
-    try:
-        if password:
-            value = getpass.getpass(display)
-        else:
-            value = input(display)
-        value = value.strip()
-        return value if value else (default or "")
-    except (KeyboardInterrupt, EOFError):
-        print()
-        return ""
-
-
-def prompt_yes_no(question: str, default: bool = True) -> bool:
-    """Prompt for a yes/no answer. Returns bool."""
-    hint = "Y/n" if default else "y/N"
-    answer = prompt(f"{question} ({hint})")
-    if not answer:
-        return default
-    return answer.lower().startswith("y")
@@ -69,8 +69,7 @@ COMMAND_REGISTRY: list[CommandDef] = [
               args_hint="[name]"),
    CommandDef("branch", "Branch the current session (explore a different path)", "Session",
               aliases=("fork",), args_hint="[name]"),
-    CommandDef("compress", "Manually compress conversation context", "Session",
-               args_hint="[focus topic]"),
+    CommandDef("compress", "Manually compress conversation context", "Session"),
    CommandDef("rollback", "List or restore filesystem checkpoints", "Session",
               args_hint="[number]"),
    CommandDef("stop", "Kill all running background processes", "Session"),
@@ -129,6 +128,7 @@ COMMAND_REGISTRY: list[CommandDef] = [
    CommandDef("cron", "Manage scheduled tasks", "Tools & Skills",
               cli_only=True, args_hint="[subcommand]",
               subcommands=("list", "add", "create", "edit", "pause", "resume", "run", "remove")),
+    CommandDef("reload", "Reload .env variables into the running session", "Tools & Skills"),
    CommandDef("reload-mcp", "Reload MCP servers from config", "Tools & Skills",
               aliases=("reload_mcp",)),
    CommandDef("browser", "Connect browser tools to your live Chrome via CDP", "Tools & Skills",
@@ -32,15 +32,13 @@ _ENV_VAR_NAME_RE = re.compile(r"^[A-Za-z_][A-Za-z0-9_]*$")
 _EXTRA_ENV_KEYS = frozenset({
    "OPENAI_API_KEY", "OPENAI_BASE_URL",
    "ANTHROPIC_API_KEY", "ANTHROPIC_TOKEN",
+    "AUXILIARY_VISION_MODEL",
    "DISCORD_HOME_CHANNEL", "TELEGRAM_HOME_CHANNEL",
    "SIGNAL_ACCOUNT", "SIGNAL_HTTP_URL",
    "SIGNAL_ALLOWED_USERS", "SIGNAL_GROUP_ALLOWED_USERS",
    "DINGTALK_CLIENT_ID", "DINGTALK_CLIENT_SECRET",
    "FEISHU_APP_ID", "FEISHU_APP_SECRET", "FEISHU_ENCRYPT_KEY", "FEISHU_VERIFICATION_TOKEN",
    "WECOM_BOT_ID", "WECOM_SECRET",
-    "WECOM_CALLBACK_CORP_ID", "WECOM_CALLBACK_CORP_SECRET", "WECOM_CALLBACK_AGENT_ID",
-    "WECOM_CALLBACK_TOKEN", "WECOM_CALLBACK_ENCODING_AES_KEY",
-    "WECOM_CALLBACK_HOST", "WECOM_CALLBACK_PORT",
    "WEIXIN_ACCOUNT_ID", "WEIXIN_TOKEN", "WEIXIN_BASE_URL", "WEIXIN_CDN_BASE_URL",
    "WEIXIN_HOME_CHANNEL", "WEIXIN_HOME_CHANNEL_NAME", "WEIXIN_DM_POLICY", "WEIXIN_GROUP_POLICY",
    "WEIXIN_ALLOWED_USERS", "WEIXIN_GROUP_ALLOWED_USERS", "WEIXIN_ALLOW_ALL_USERS",
@@ -143,73 +141,6 @@ def managed_error(action: str = "modify configuration"):
    print(format_managed_message(action), file=sys.stderr)


-# =============================================================================
-# Container-aware CLI (NixOS container mode)
-# =============================================================================
-
-def _is_inside_container() -> bool:
-    """Detect if we're already running inside a Docker/Podman container."""
-    # Standard Docker/Podman indicators
-    if os.path.exists("/.dockerenv"):
-        return True
-    # Podman uses /run/.containerenv
-    if os.path.exists("/run/.containerenv"):
-        return True
-    # Check cgroup for container runtime evidence (works for both Docker & Podman)
-    try:
-        with open("/proc/1/cgroup", "r") as f:
-            cgroup = f.read()
-            if "docker" in cgroup or "podman" in cgroup or "/lxc/" in cgroup:
-                return True
-    except OSError:
-        pass
-    return False
-
-
-def get_container_exec_info() -> Optional[dict]:
-    """Read container mode metadata from HERMES_HOME/.container-mode.
-
-    Returns a dict with keys: backend, container_name, exec_user, hermes_bin
-    or None if container mode is not active, we're already inside the
-    container, or HERMES_DEV=1 is set.
-
-    The .container-mode file is written by the NixOS activation script when
-    container.enable = true. It tells the host CLI to exec into the container
-    instead of running locally.
-    """
-    if os.environ.get("HERMES_DEV") == "1":
-        return None
-
-    if _is_inside_container():
-        return None
-
-    container_mode_file = get_hermes_home() / ".container-mode"
-
-    try:
-        info = {}
-        with open(container_mode_file, "r") as f:
-            for line in f:
-                line = line.strip()
-                if "=" in line and not line.startswith("#"):
-                    key, _, value = line.partition("=")
-                    info[key.strip()] = value.strip()
-    except FileNotFoundError:
-        return None
-    # All other exceptions (PermissionError, malformed data, etc.) propagate
-
-    backend = info.get("backend", "docker")
-    container_name = info.get("container_name", "hermes-agent")
-    exec_user = info.get("exec_user", "hermes")
-    hermes_bin = info.get("hermes_bin", "/data/current-package/bin/hermes")
-
-    return {
-        "backend": backend,
-        "container_name": container_name,
-        "exec_user": exec_user,
-        "hermes_bin": hermes_bin,
-    }
-
-
 # =============================================================================
 # Config paths
 # =============================================================================
@@ -450,7 +381,7 @@ DEFAULT_CONFIG = {
            "model": "",           # e.g. "google/gemini-2.5-flash", "gpt-4o"
            "base_url": "",        # direct OpenAI-compatible endpoint (takes precedence over provider)
            "api_key": "",         # API key for base_url (falls back to OPENAI_API_KEY)
-            "timeout": 120,        # seconds — LLM API call timeout; vision payloads need generous timeout
+            "timeout": 30,         # seconds — LLM API call timeout; increase for slow local vision models
            "download_timeout": 30,  # seconds — image HTTP download timeout; increase for slow connections
        },
        "web_extract": {
@@ -515,11 +446,9 @@ DEFAULT_CONFIG = {
        "inline_diffs": True,     # Show inline diff previews for write actions (write_file, patch, skill_manage)
        "show_cost": False,       # Show $ cost in the status bar (off by default)
        "skin": "default",
-        "interim_assistant_messages": True,  # Gateway: show natural mid-turn assistant status messages
        "tool_progress_command": False,  # Enable /verbose command in messaging gateway
-        "tool_progress_overrides": {},  # DEPRECATED — use display.platforms instead
+        "tool_progress_overrides": {},  # Per-platform overrides: {"signal": "off", "telegram": "all"}
        "tool_preview_length": 0,  # Max chars for tool call previews (0 = no limit, show full paths/commands)
-        "platforms": {},  # Per-platform display overrides: {"telegram": {"tool_progress": "all"}, "slack": {"tool_progress": "off"}}
    },

    # Privacy settings
@@ -529,7 +458,7 @@ DEFAULT_CONFIG = {
    
    # Text-to-speech configuration
    "tts": {
-        "provider": "edge",  # "edge" (free) | "elevenlabs" (premium) | "openai" | "minimax" | "mistral" | "neutts" (local)
+        "provider": "edge",  # "edge" (free) | "elevenlabs" (premium) | "openai" | "neutts" (local)
        "edge": {
            "voice": "en-US-AriaNeural",
            # Popular: AriaNeural, JennyNeural, AndrewNeural, BrianNeural, SoniaNeural
@@ -543,10 +472,6 @@ DEFAULT_CONFIG = {
            "voice": "alloy",
            # Voices: alloy, echo, fable, onyx, nova, shimmer
        },
-        "mistral": {
-            "model": "voxtral-mini-tts-2603",
-            "voice_id": "c69964a6-ab8b-4f8a-9465-ec0925096ec8",  # Paul - Neutral
-        },
        "neutts": {
            "ref_audio": "",  # Path to reference voice audio (empty = bundled default)
            "ref_text": "",   # Path to reference voice transcript (empty = bundled default)
@@ -706,16 +631,8 @@ DEFAULT_CONFIG = {
        "backup_count": 3,     # Number of rotated backup files to keep
    },

-    # Network settings — workarounds for connectivity issues.
-    "network": {
-        # Force IPv4 connections.  On servers with broken or unreachable IPv6,
-        # Python tries AAAA records first and hangs for the full TCP timeout
-        # before falling back to IPv4.  Set to true to skip IPv6 entirely.
-        "force_ipv4": False,
-    },
-
    # Config schema version - bump this when adding new required fields
-    "_config_version": 16,
+    "_config_version": 14,
 }

 # =============================================================================
@@ -947,21 +864,6 @@ OPTIONAL_ENV_VARS = {
        "category": "provider",
        "advanced": True,
    },
-    "XIAOMI_API_KEY": {
-        "description": "Xiaomi MiMo API key for MiMo models (mimo-v2-pro, mimo-v2-omni, mimo-v2-flash)",
-        "prompt": "Xiaomi MiMo API Key",
-        "url": "https://platform.xiaomimimo.com",
-        "password": True,
-        "category": "provider",
-    },
-    "XIAOMI_BASE_URL": {
-        "description": "Xiaomi MiMo base URL override (default: https://api.xiaomimimo.com/v1)",
-        "prompt": "Xiaomi base URL (leave empty for default)",
-        "url": None,
-        "password": False,
-        "category": "provider",
-        "advanced": True,
-    },

    # ── Tool API keys ──
    "EXA_API_KEY": {
@@ -1114,13 +1016,6 @@ OPTIONAL_ENV_VARS = {
        "password": True,
        "category": "tool",
    },
-    "MISTRAL_API_KEY": {
-        "description": "Mistral API key for Voxtral TTS and transcription (STT)",
-        "prompt": "Mistral API key",
-        "url": "https://console.mistral.ai/",
-        "password": True,
-        "category": "tool",
-    },
    "GITHUB_TOKEN": {
        "description": "GitHub token for Skills Hub (higher API rate limits, skill publish)",
        "prompt": "GitHub Token",
@@ -1577,7 +1472,7 @@ _KNOWN_ROOT_KEYS = {

 # Valid fields inside a custom_providers list entry
 _VALID_CUSTOM_PROVIDER_FIELDS = {
-    "name", "base_url", "api_key", "api_mode", "model", "models",
+    "name", "base_url", "api_key", "api_mode", "models",
    "context_length", "rate_limit_delay",
 }

@@ -1942,44 +1837,6 @@ def migrate_config(interactive: bool = True, quiet: bool = False) -> Dict[str, A
            if not quiet:
                print(f"  ✓ Migrated legacy stt.model to provider-specific config")

-    # ── Version 14 → 15: add explicit gateway interim-message gate ──
-    if current_ver < 15:
-        config = read_raw_config()
-        display = config.get("display", {})
-        if not isinstance(display, dict):
-            display = {}
-        if "interim_assistant_messages" not in display:
-            display["interim_assistant_messages"] = True
-            config["display"] = display
-            results["config_added"].append("display.interim_assistant_messages=true (default)")
-            save_config(config)
-            if not quiet:
-                print("  ✓ Added display.interim_assistant_messages=true")
-
-    # ── Version 15 → 16: migrate tool_progress_overrides into display.platforms ──
-    if current_ver < 16:
-        config = read_raw_config()
-        display = config.get("display", {})
-        if not isinstance(display, dict):
-            display = {}
-        old_overrides = display.get("tool_progress_overrides")
-        if isinstance(old_overrides, dict) and old_overrides:
-            platforms = display.get("platforms", {})
-            if not isinstance(platforms, dict):
-                platforms = {}
-            for plat, mode in old_overrides.items():
-                if plat not in platforms:
-                    platforms[plat] = {}
-                if "tool_progress" not in platforms[plat]:
-                    platforms[plat]["tool_progress"] = mode
-            display["platforms"] = platforms
-            config["display"] = display
-            save_config(config)
-            if not quiet:
-                migrated = ", ".join(f"{p}={m}" for p, m in old_overrides.items())
-                print(f"  ✓ Migrated tool_progress_overrides → display.platforms: {migrated}")
-            results["config_added"].append("display.platforms (migrated from tool_progress_overrides)")
-
    if current_ver < latest_ver and not quiet:
        print(f"Config version: {current_ver} → {latest_ver}")
    
@@ -2640,6 +2497,51 @@ def save_env_value_secure(key: str, value: str) -> Dict[str, Any]:
    }


+def delete_env_value(key: str) -> bool:
+    """Remove a key from ~/.hermes/.env. Returns True if the key was found and removed."""
+    env_path = get_env_path()
+    if not env_path.exists():
+        return False
+
+    read_kw = {"encoding": "utf-8", "errors": "replace"} if _IS_WINDOWS else {}
+    write_kw = {"encoding": "utf-8"} if _IS_WINDOWS else {}
+
+    with open(env_path, **read_kw) as f:
+        lines = f.readlines()
+
+    new_lines = [l for l in lines if not l.strip().startswith(f"{key}=")]
+    if len(new_lines) == len(lines):
+        return False
+
+    fd, tmp_path = tempfile.mkstemp(dir=str(env_path.parent), suffix='.tmp', prefix='.env_')
+    try:
+        with os.fdopen(fd, 'w', **write_kw) as f:
+            f.writelines(new_lines)
+            f.flush()
+            os.fsync(f.fileno())
+        os.replace(tmp_path, env_path)
+    except BaseException:
+        try:
+            os.unlink(tmp_path)
+        except OSError:
+            pass
+        raise
+    _secure_file(env_path)
+
+    os.environ.pop(key, None)
+    return True
+
+
+def reload_env() -> int:
+    """Re-read ~/.hermes/.env into os.environ. Returns count of vars updated."""
+    env_vars = load_env()
+    count = 0
+    for key, value in env_vars.items():
+        if os.environ.get(key) != value:
+            os.environ[key] = value
+            count += 1
+    return count
+

 def get_env_value(key: str) -> Optional[str]:
    """Get a value from ~/.hermes/.env or environment."""
@@ -2700,8 +2602,7 @@ def show_config():
    for env_key, name in keys:
        value = get_env_value(env_key)
        print(f"  {name:<14} {redact_key(value)}")
-    from hermes_cli.auth import get_anthropic_key
-    anthropic_value = get_anthropic_key()
+    anthropic_value = get_env_value("ANTHROPIC_TOKEN") or get_env_value("ANTHROPIC_API_KEY")
    print(f"  {'Anthropic':<14} {redact_key(anthropic_value)}")
    
    # Model settings
@@ -2917,8 +2818,8 @@ def set_config_value(key: str, value: str):
    
    # Write only user config back (not the full merged defaults)
    ensure_hermes_home()
-    from utils import atomic_yaml_write
-    atomic_yaml_write(config_path, user_config, sort_keys=False)
+    with open(config_path, 'w', encoding="utf-8") as f:
+        yaml.dump(user_config, f, default_flow_style=False, sort_keys=False)
    
    # Keep .env in sync for keys that terminal_tool reads directly from env vars.
    # config.yaml is authoritative, but terminal_tool only reads TERMINAL_ENV etc.
@@ -287,129 +287,6 @@ def _radio_numbered_fallback(
        return cancel_returns


-def curses_single_select(
-    title: str,
-    items: List[str],
-    default_index: int = 0,
-    *,
-    cancel_label: str = "Cancel",
-) -> int | None:
-    """Curses single-select menu. Returns selected index or None on cancel.
-
-    Works inside prompt_toolkit because curses.wrapper() restores the terminal
-    safely, unlike simple_term_menu which conflicts with /dev/tty.
-    """
-    if not sys.stdin.isatty():
-        return None
-
-    try:
-        import curses
-        result_holder: list = [None]
-
-        all_items = list(items) + [cancel_label]
-        cancel_idx = len(items)
-
-        def _draw(stdscr):
-            curses.curs_set(0)
-            if curses.has_colors():
-                curses.start_color()
-                curses.use_default_colors()
-                curses.init_pair(1, curses.COLOR_GREEN, -1)
-                curses.init_pair(2, curses.COLOR_YELLOW, -1)
-            cursor = min(default_index, len(all_items) - 1)
-            scroll_offset = 0
-
-            while True:
-                stdscr.clear()
-                max_y, max_x = stdscr.getmaxyx()
-
-                try:
-                    hattr = curses.A_BOLD
-                    if curses.has_colors():
-                        hattr |= curses.color_pair(2)
-                    stdscr.addnstr(0, 0, title, max_x - 1, hattr)
-                    stdscr.addnstr(
-                        1, 0,
-                        "  ↑↓ navigate  ENTER confirm  ESC/q cancel",
-                        max_x - 1, curses.A_DIM,
-                    )
-                except curses.error:
-                    pass
-
-                visible_rows = max_y - 3
-                if cursor < scroll_offset:
-                    scroll_offset = cursor
-                elif cursor >= scroll_offset + visible_rows:
-                    scroll_offset = cursor - visible_rows + 1
-
-                for draw_i, i in enumerate(
-                    range(scroll_offset, min(len(all_items), scroll_offset + visible_rows))
-                ):
-                    y = draw_i + 3
-                    if y >= max_y - 1:
-                        break
-                    arrow = "→" if i == cursor else " "
-                    line = f" {arrow} {all_items[i]}"
-                    attr = curses.A_NORMAL
-                    if i == cursor:
-                        attr = curses.A_BOLD
-                        if curses.has_colors():
-                            attr |= curses.color_pair(1)
-                    try:
-                        stdscr.addnstr(y, 0, line, max_x - 1, attr)
-                    except curses.error:
-                        pass
-
-                stdscr.refresh()
-                key = stdscr.getch()
-
-                if key in (curses.KEY_UP, ord("k")):
-                    cursor = (cursor - 1) % len(all_items)
-                elif key in (curses.KEY_DOWN, ord("j")):
-                    cursor = (cursor + 1) % len(all_items)
-                elif key in (curses.KEY_ENTER, 10, 13):
-                    result_holder[0] = cursor
-                    return
-                elif key in (27, ord("q")):
-                    result_holder[0] = None
-                    return
-
-        curses.wrapper(_draw)
-        flush_stdin()
-        if result_holder[0] is not None and result_holder[0] >= cancel_idx:
-            return None
-        return result_holder[0]
-
-    except Exception:
-        all_items = list(items) + [cancel_label]
-        cancel_idx = len(items)
-        return _numbered_single_fallback(title, all_items, cancel_idx)
-
-
-def _numbered_single_fallback(
-    title: str,
-    items: List[str],
-    cancel_idx: int,
-) -> int | None:
-    """Text-based numbered fallback for single-select."""
-    print(f"\n  {title}\n")
-    for i, label in enumerate(items, 1):
-        print(f"  {i}. {label}")
-    print()
-    try:
-        val = input(f"  Choice [1-{len(items)}]: ").strip()
-        if not val:
-            return None
-        idx = int(val) - 1
-        if 0 <= idx < len(items) and idx < cancel_idx:
-            return idx
-        if idx == cancel_idx:
-            return None
-    except (ValueError, KeyboardInterrupt, EOFError):
-        pass
-    return None
-
-
 def _numbered_fallback(
    title: str,
    items: List[str],
@@ -51,7 +51,6 @@ _PROVIDER_ENV_HINTS = (
    "AI_GATEWAY_API_KEY",
    "OPENCODE_ZEN_API_KEY",
    "OPENCODE_GO_API_KEY",
-    "XIAOMI_API_KEY",
 )


@@ -336,8 +335,8 @@ def run_doctor(args):
                            model_section[k] = raw_config.pop(k)
                        else:
                            raw_config.pop(k)
-                    from utils import atomic_yaml_write
-                    atomic_yaml_write(config_path, raw_config)
+                    with open(config_path, "w") as f:
+                        yaml.dump(raw_config, f, default_flow_style=False)
                    check_ok("Migrated stale root-level keys into model section")
                    fixed_count += 1
                else:
@@ -686,8 +685,7 @@ def run_doctor(args):
    else:
        check_warn("OpenRouter API", "(not configured)")
    
-    from hermes_cli.auth import get_anthropic_key
-    anthropic_key = get_anthropic_key()
+    anthropic_key = os.getenv("ANTHROPIC_TOKEN") or os.getenv("ANTHROPIC_API_KEY")
    if anthropic_key:
        print("  Checking Anthropic API...", end="", flush=True)
        try:
@@ -724,9 +722,9 @@ def run_doctor(args):
        ("DeepSeek",         ("DEEPSEEK_API_KEY",),                           "https://api.deepseek.com/v1/models",  "DEEPSEEK_BASE_URL", True),
        ("Hugging Face",     ("HF_TOKEN",),                                   "https://router.huggingface.co/v1/models", "HF_BASE_URL", True),
        ("Alibaba/DashScope", ("DASHSCOPE_API_KEY",),                         "https://dashscope-intl.aliyuncs.com/compatible-mode/v1/models", "DASHSCOPE_BASE_URL", True),
-        # MiniMax: the /anthropic endpoint doesn't support /models, but the /v1 endpoint does.
-        ("MiniMax",          ("MINIMAX_API_KEY",),                            "https://api.minimax.io/v1/models",    "MINIMAX_BASE_URL", True),
-        ("MiniMax (China)",  ("MINIMAX_CN_API_KEY",),                         "https://api.minimaxi.com/v1/models",  "MINIMAX_CN_BASE_URL", True),
+        # MiniMax APIs don't support /models endpoint — https://github.com/NousResearch/hermes-agent/issues/811
+        ("MiniMax",          ("MINIMAX_API_KEY",),                            None,                                  "MINIMAX_BASE_URL", False),
+        ("MiniMax (China)",  ("MINIMAX_CN_API_KEY",),                         None,                                  "MINIMAX_CN_BASE_URL", False),
        ("AI Gateway",       ("AI_GATEWAY_API_KEY",),                          "https://ai-gateway.vercel.sh/v1/models", "AI_GATEWAY_BASE_URL", True),
        ("Kilo Code",        ("KILOCODE_API_KEY",),                            "https://api.kilo.ai/api/gateway/models",  "KILOCODE_BASE_URL", True),
        ("OpenCode Zen",     ("OPENCODE_ZEN_API_KEY",),                        "https://opencode.ai/zen/v1/models",  "OPENCODE_ZEN_BASE_URL", True),
@@ -751,11 +749,6 @@ def run_doctor(args):
                # Auto-detect Kimi Code keys (sk-kimi-) → api.kimi.com
                if not _base and _key.startswith("sk-kimi-"):
                    _base = "https://api.kimi.com/coding/v1"
-                # Anthropic-compat endpoints (/anthropic) don't support /models.
-                # Rewrite to the OpenAI-compat /v1 surface for health checks.
-                if _base and _base.rstrip("/").endswith("/anthropic"):
-                    from agent.auxiliary_client import _to_openai_base_url
-                    _base = _to_openai_base_url(_base)
                _url = (_base.rstrip("/") + "/models") if _base else _default_url
                _headers = {"Authorization": f"Bearer {_key}"}
                if "api.kimi.com" in _url.lower():
@@ -119,7 +119,6 @@ def _configured_platforms() -> list[str]:
        "dingtalk": "DINGTALK_CLIENT_ID",
        "feishu": "FEISHU_APP_ID",
        "wecom": "WECOM_BOT_ID",
-        "wecom_callback": "WECOM_CALLBACK_CORP_ID",
        "weixin": "WEIXIN_ACCOUNT_ID",
    }
    return [name for name, env in checks.items() if os.getenv(env)]
@@ -157,54 +157,30 @@ def _request_gateway_self_restart(pid: int) -> bool:
    return True


-def find_gateway_pids(exclude_pids: set | None = None, all_profiles: bool = False) -> list:
+def find_gateway_pids(exclude_pids: set | None = None) -> list:
    """Find PIDs of running gateway processes.

    Args:
        exclude_pids: PIDs to exclude from the result (e.g. service-managed
            PIDs that should not be killed during a stale-process sweep).
-        all_profiles: When ``True``, return gateway PIDs across **all**
-            profiles (the pre-7923 global behaviour).  ``hermes update``
-            needs this because a code update affects every profile.
-            When ``False`` (default), only PIDs belonging to the current
-            Hermes profile are returned.
    """
+    pids = []
    _exclude = exclude_pids or set()
-    pids = [pid for pid in _get_service_pids() if pid not in _exclude]
    patterns = [
        "hermes_cli.main gateway",
-        "hermes_cli.main --profile",
-        "hermes_cli.main -p",
        "hermes_cli/main.py gateway",
-        "hermes_cli/main.py --profile",
-        "hermes_cli/main.py -p",
        "hermes gateway",
        "gateway/run.py",
    ]
-    current_home = str(get_hermes_home().resolve())
-    current_profile_arg = _profile_arg(current_home)
-    current_profile_name = current_profile_arg.split()[-1] if current_profile_arg else ""
-
-    def _matches_current_profile(command: str) -> bool:
-        if current_profile_name:
-            return (
-                f"--profile {current_profile_name}" in command
-                or f"-p {current_profile_name}" in command
-                or f"HERMES_HOME={current_home}" in command
-            )
-
-        if "--profile " in command or " -p " in command:
-            return False
-        if "HERMES_HOME=" in command and f"HERMES_HOME={current_home}" not in command:
-            return False
-        return True

    try:
        if is_windows():
+            # Windows: use wmic to search command lines
            result = subprocess.run(
                ["wmic", "process", "get", "ProcessId,CommandLine", "/FORMAT:LIST"],
                capture_output=True, text=True, timeout=10
            )
+            # Parse WMIC LIST output: blocks of "CommandLine=...\nProcessId=...\n"
            current_cmd = ""
            for line in result.stdout.split('\n'):
                line = line.strip()
@@ -212,7 +188,7 @@ def find_gateway_pids(exclude_pids: set | None = None, all_profiles: bool = Fals
                    current_cmd = line[len("CommandLine="):]
                elif line.startswith("ProcessId="):
                    pid_str = line[len("ProcessId="):]
-                    if any(p in current_cmd for p in patterns) and (all_profiles or _matches_current_profile(current_cmd)):
+                    if any(p in current_cmd for p in patterns):
                        try:
                            pid = int(pid_str)
                            if pid != os.getpid() and pid not in pids and pid not in _exclude:
@@ -222,57 +198,41 @@ def find_gateway_pids(exclude_pids: set | None = None, all_profiles: bool = Fals
                    current_cmd = ""
        else:
            result = subprocess.run(
-                ["ps", "eww", "-ax", "-o", "pid=,command="],
+                ["ps", "aux"],
                capture_output=True,
                text=True,
                timeout=10,
            )
            for line in result.stdout.split('\n'):
-                stripped = line.strip()
-                if not stripped or 'grep' in stripped:
+                # Skip grep and current process
+                if 'grep' in line or str(os.getpid()) in line:
                    continue
-
-                pid = None
-                command = ""
-
-                parts = stripped.split(None, 1)
-                if len(parts) == 2:
-                    try:
-                        pid = int(parts[0])
-                        command = parts[1]
-                    except ValueError:
-                        pid = None
-
-                if pid is None:
-                    aux_parts = stripped.split()
-                    if len(aux_parts) > 10 and aux_parts[1].isdigit():
-                        pid = int(aux_parts[1])
-                        command = " ".join(aux_parts[10:])
-
-                if pid is None:
-                    continue
-                if pid == os.getpid() or pid in pids or pid in _exclude:
-                    continue
-                if any(pattern in command for pattern in patterns) and (all_profiles or _matches_current_profile(command)):
-                    pids.append(pid)
-    except (OSError, subprocess.TimeoutExpired):
+                for pattern in patterns:
+                    if pattern in line:
+                        parts = line.split()
+                        if len(parts) > 1:
+                            try:
+                                pid = int(parts[1])
+                                if pid not in pids and pid not in _exclude:
+                                    pids.append(pid)
+                            except ValueError:
+                                continue
+                        break
+    except Exception:
        pass

    return pids


-def kill_gateway_processes(force: bool = False, exclude_pids: set | None = None,
-                           all_profiles: bool = False) -> int:
+def kill_gateway_processes(force: bool = False, exclude_pids: set | None = None) -> int:
    """Kill any running gateway processes. Returns count killed.

    Args:
        force: Use the platform's force-kill mechanism instead of graceful terminate.
        exclude_pids: PIDs to skip (e.g. service-managed PIDs that were just
            restarted and should not be killed).
-        all_profiles: When ``True``, kill across all profiles.  Passed
-            through to :func:`find_gateway_pids`.
    """
-    pids = find_gateway_pids(exclude_pids=exclude_pids, all_profiles=all_profiles)
+    pids = find_gateway_pids(exclude_pids=exclude_pids)
    killed = 0
    
    for pid in pids:
@@ -673,17 +633,6 @@ def print_systemd_linger_guidance() -> None:
        print("  If you want the gateway user service to survive logout, run:")
        print("  sudo loginctl enable-linger $USER")

-def _launchd_user_home() -> Path:
-    """Return the real macOS user home for launchd artifacts.
-
-    Profile-mode Hermes often sets ``HOME`` to a profile-scoped directory, but
-    launchd user agents still live under the actual account home.
-    """
-    import pwd
-
-    return Path(pwd.getpwuid(os.getuid()).pw_dir)
-
-
 def get_launchd_plist_path() -> Path:
    """Return the launchd plist path, scoped per profile.

@@ -692,7 +641,7 @@ def get_launchd_plist_path() -> Path:
    """
    suffix = _profile_suffix()
    name = f"ai.hermes.gateway-{suffix}" if suffix else "ai.hermes.gateway"
-    return _launchd_user_home() / "Library" / "LaunchAgents" / f"{name}.plist"
+    return Path.home() / "Library" / "LaunchAgents" / f"{name}.plist"

 def _detect_venv_dir() -> Path | None:
    """Detect the active virtualenv directory.
@@ -890,25 +839,6 @@ def _normalize_service_definition(text: str) -> str:
    return "\n".join(line.rstrip() for line in text.strip().splitlines())


-def _normalize_launchd_plist_for_comparison(text: str) -> str:
-    """Normalize launchd plist text for staleness checks.
-
-    The generated plist intentionally captures a broad PATH assembled from the
-    invoking shell so user-installed tools remain reachable under launchd.
-    That makes raw text comparison unstable across shells, so ignore the PATH
-    payload when deciding whether the installed plist is stale.
-    """
-    import re
-
-    normalized = _normalize_service_definition(text)
-    return re.sub(
-        r'(<key>PATH</key>\s*<string>)(.*?)(</string>)',
-        r'\1__HERMES_PATH__\3',
-        normalized,
-        flags=re.S,
-    )
-
-
 def systemd_unit_is_current(system: bool = False) -> bool:
    unit_path = get_systemd_unit_path(system=system)
    if not unit_path.exists():
@@ -1290,7 +1220,7 @@ def launchd_plist_is_current() -> bool:

    installed = plist_path.read_text(encoding="utf-8")
    expected = generate_launchd_plist()
-    return _normalize_launchd_plist_for_comparison(installed) == _normalize_launchd_plist_for_comparison(expected)
+    return _normalize_service_definition(installed) == _normalize_service_definition(expected)


 def refresh_launchd_plist_if_needed() -> bool:
@@ -1821,37 +1751,6 @@ _PLATFORMS = [
             "help": "Chat ID for scheduled results and notifications."},
        ],
    },
-    {
-        "key": "wecom_callback",
-        "label": "WeCom Callback (Self-Built App)",
-        "emoji": "💬",
-        "token_var": "WECOM_CALLBACK_CORP_ID",
-        "setup_instructions": [
-            "1. Go to WeCom Admin Console → Applications → Create Self-Built App",
-            "2. Note the Corp ID (top of admin console) and create a Corp Secret",
-            "3. Under Receive Messages, configure the callback URL to point to your server",
-            "4. Copy the Token and EncodingAESKey from the callback configuration",
-            "5. The adapter runs an HTTP server — ensure the port is reachable from WeCom",
-            "6. Restrict access with WECOM_CALLBACK_ALLOWED_USERS for production use",
-        ],
-        "vars": [
-            {"name": "WECOM_CALLBACK_CORP_ID", "prompt": "Corp ID", "password": False,
-             "help": "Your WeCom enterprise Corp ID."},
-            {"name": "WECOM_CALLBACK_CORP_SECRET", "prompt": "Corp Secret", "password": True,
-             "help": "The secret for your self-built application."},
-            {"name": "WECOM_CALLBACK_AGENT_ID", "prompt": "Agent ID", "password": False,
-             "help": "The Agent ID of your self-built application."},
-            {"name": "WECOM_CALLBACK_TOKEN", "prompt": "Callback Token", "password": True,
-             "help": "The Token from your WeCom callback configuration."},
-            {"name": "WECOM_CALLBACK_ENCODING_AES_KEY", "prompt": "Encoding AES Key", "password": True,
-             "help": "The EncodingAESKey from your WeCom callback configuration."},
-            {"name": "WECOM_CALLBACK_PORT", "prompt": "Callback server port (default: 8645)", "password": False,
-             "help": "Port for the HTTP callback server."},
-            {"name": "WECOM_CALLBACK_ALLOWED_USERS", "prompt": "Allowed user IDs (comma-separated, or empty)", "password": False,
-             "is_allowlist": True,
-             "help": "Restrict which WeCom users can interact with the app."},
-        ],
-    },
    {
        "key": "weixin",
        "label": "Weixin / WeChat",
@@ -2082,36 +1981,6 @@ def _setup_whatsapp():
    cmd_whatsapp(argparse.Namespace())


-def _setup_email():
-    """Configure Email via the standard platform setup."""
-    email_platform = next(p for p in _PLATFORMS if p["key"] == "email")
-    _setup_standard_platform(email_platform)
-
-
-def _setup_sms():
-    """Configure SMS (Twilio) via the standard platform setup."""
-    sms_platform = next(p for p in _PLATFORMS if p["key"] == "sms")
-    _setup_standard_platform(sms_platform)
-
-
-def _setup_dingtalk():
-    """Configure DingTalk via the standard platform setup."""
-    dingtalk_platform = next(p for p in _PLATFORMS if p["key"] == "dingtalk")
-    _setup_standard_platform(dingtalk_platform)
-
-
-def _setup_feishu():
-    """Configure Feishu / Lark via the standard platform setup."""
-    feishu_platform = next(p for p in _PLATFORMS if p["key"] == "feishu")
-    _setup_standard_platform(feishu_platform)
-
-
-def _setup_wecom():
-    """Configure WeCom (Enterprise WeChat) via the standard platform setup."""
-    wecom_platform = next(p for p in _PLATFORMS if p["key"] == "wecom")
-    _setup_standard_platform(wecom_platform)
-
-
 def _is_service_installed() -> bool:
    """Check if the gateway is installed as a system service."""
    if supports_systemd_services():
@@ -2671,7 +2540,7 @@ def gateway_command(args):
                    service_available = True
                except subprocess.CalledProcessError:
                    pass
-            killed = kill_gateway_processes(all_profiles=True)
+            killed = kill_gateway_processes()
            total = killed + (1 if service_available else 0)
            if total:
                print(f"✓ Stopped {total} gateway process(es) across all profiles")
@@ -1,18 +1,16 @@
 """``hermes logs`` — view and filter Hermes log files.

-Supports tailing, following, session filtering, level filtering,
-component filtering, and relative time ranges.  All log files live
-under ``~/.hermes/logs/``.
+Supports tailing, following, session filtering, level filtering, and
+relative time ranges.  All log files live under ``~/.hermes/logs/``.

 Usage examples::

    hermes logs                    # last 50 lines of agent.log
    hermes logs -f                 # follow agent.log in real time
    hermes logs errors             # last 50 lines of errors.log
-    hermes logs gateway -n 100    # last 100 lines of gateway.log
+    hermes logs gateway -n 100     # last 100 lines of gateway.log
    hermes logs --level WARNING    # only WARNING+ lines
    hermes logs --session abc123   # filter by session ID substring
-    hermes logs --component tools  # only tool-related lines
    hermes logs --since 1h         # lines from the last hour
    hermes logs --since 30m -f     # follow, starting 30 min ago
 """
@@ -22,7 +20,7 @@ import sys
 import time
 from datetime import datetime, timedelta
 from pathlib import Path
-from typing import Optional, Sequence
+from typing import Optional

 from hermes_constants import get_hermes_home, display_hermes_home

@@ -40,15 +38,6 @@ _TS_RE = re.compile(r"^(\d{4}-\d{2}-\d{2}\s+\d{2}:\d{2}:\d{2})")
 # Level extraction — matches " INFO ", " WARNING ", " ERROR ", " DEBUG ", " CRITICAL "
 _LEVEL_RE = re.compile(r"\s(DEBUG|INFO|WARNING|ERROR|CRITICAL)\s")

-# Logger name extraction — after level and optional session tag, the next
-# non-space token before ":" is the logger name.
-# Matches: "INFO gateway.run:" or "INFO [sess_abc] tools.terminal_tool:"
-_LOGGER_NAME_RE = re.compile(
-    r"\s(?:DEBUG|INFO|WARNING|ERROR|CRITICAL)"  # level
-    r"(?:\s+\[.*?\])?"                           # optional session tag
-    r"\s+(\S+):"                                 # logger name
-)
-
 # Level ordering for >= filtering
 _LEVEL_ORDER = {"DEBUG": 0, "INFO": 1, "WARNING": 2, "ERROR": 3, "CRITICAL": 4}

@@ -90,27 +79,12 @@ def _extract_level(line: str) -> Optional[str]:
    return m.group(1) if m else None


-def _extract_logger_name(line: str) -> Optional[str]:
-    """Extract the logger name from a log line."""
-    m = _LOGGER_NAME_RE.search(line)
-    return m.group(1) if m else None
-
-
-def _line_matches_component(line: str, prefixes: Sequence[str]) -> bool:
-    """Check if a log line's logger name starts with any of *prefixes*."""
-    name = _extract_logger_name(line)
-    if name is None:
-        return False
-    return name.startswith(tuple(prefixes))
-
-
 def _matches_filters(
    line: str,
    *,
    min_level: Optional[str] = None,
    session_filter: Optional[str] = None,
    since: Optional[datetime] = None,
-    component_prefixes: Optional[Sequence[str]] = None,
 ) -> bool:
    """Check if a log line passes all active filters."""
    if since is not None:
@@ -128,10 +102,6 @@ def _matches_filters(
        if session_filter not in line:
            return False

-    if component_prefixes is not None:
-        if not _line_matches_component(line, component_prefixes):
-            return False
-
    return True


@@ -143,7 +113,6 @@ def tail_log(
    level: Optional[str] = None,
    session: Optional[str] = None,
    since: Optional[str] = None,
-    component: Optional[str] = None,
 ) -> None:
    """Read and display log lines, optionally following in real time.

@@ -161,8 +130,6 @@ def tail_log(
        Session ID substring to filter on.
    since
        Relative time string (e.g. ``"1h"``, ``"30m"``).
-    component
-        Component name to filter by (e.g. ``"gateway"``, ``"tools"``).
    """
    filename = LOG_FILES.get(log_name)
    if filename is None:
@@ -188,29 +155,13 @@ def tail_log(
        print(f"Invalid --level: {level!r}. Use DEBUG, INFO, WARNING, ERROR, or CRITICAL.")
        sys.exit(1)

-    # Resolve component to logger name prefixes
-    component_prefixes = None
-    if component:
-        from hermes_logging import COMPONENT_PREFIXES
-        component_lower = component.lower()
-        if component_lower not in COMPONENT_PREFIXES:
-            available = ", ".join(sorted(COMPONENT_PREFIXES))
-            print(f"Unknown component: {component!r}. Available: {available}")
-            sys.exit(1)
-        component_prefixes = COMPONENT_PREFIXES[component_lower]
-
-    has_filters = (
-        min_level is not None
-        or session is not None
-        or since_dt is not None
-        or component_prefixes is not None
-    )
+    has_filters = min_level is not None or session is not None or since_dt is not None

    # Read and display the tail
    try:
        lines = _read_tail(log_path, num_lines, has_filters=has_filters,
                           min_level=min_level, session_filter=session,
-                           since=since_dt, component_prefixes=component_prefixes)
+                           since=since_dt)
    except PermissionError:
        print(f"Permission denied: {log_path}")
        sys.exit(1)
@@ -221,8 +172,6 @@ def tail_log(
        filter_parts.append(f"level>={min_level}")
    if session:
        filter_parts.append(f"session={session}")
-    if component:
-        filter_parts.append(f"component={component}")
    if since:
        filter_parts.append(f"since={since}")
    filter_desc = f" [{', '.join(filter_parts)}]" if filter_parts else ""
@@ -241,7 +190,7 @@ def tail_log(
    # Follow mode — poll for new content
    try:
        _follow_log(log_path, min_level=min_level, session_filter=session,
-                     since=since_dt, component_prefixes=component_prefixes)
+                     since=since_dt)
    except KeyboardInterrupt:
        print("\n--- stopped ---")

@@ -254,7 +203,6 @@ def _read_tail(
    min_level: Optional[str] = None,
    session_filter: Optional[str] = None,
    since: Optional[datetime] = None,
-    component_prefixes: Optional[Sequence[str]] = None,
 ) -> list:
    """Read the last *num_lines* matching lines from a log file.

@@ -267,8 +215,7 @@ def _read_tail(
        filtered = [
            l for l in raw_lines
            if _matches_filters(l, min_level=min_level,
-                                session_filter=session_filter, since=since,
-                                component_prefixes=component_prefixes)
+                                session_filter=session_filter, since=since)
        ]
        return filtered[-num_lines:]
    else:
@@ -337,7 +284,6 @@ def _follow_log(
    min_level: Optional[str] = None,
    session_filter: Optional[str] = None,
    since: Optional[datetime] = None,
-    component_prefixes: Optional[Sequence[str]] = None,
 ) -> None:
    """Poll a log file for new content and print matching lines."""
    with open(path, "r", encoding="utf-8", errors="replace") as f:
@@ -347,8 +293,7 @@ def _follow_log(
            line = f.readline()
            if line:
                if _matches_filters(line, min_level=min_level,
-                                    session_filter=session_filter, since=since,
-                                    component_prefixes=component_prefixes):
+                                    session_filter=session_filter, since=since):
                    print(line, end="")
                    sys.stdout.flush()
            else:
@@ -151,18 +151,6 @@ try:
 except Exception:
    pass  # best-effort — don't crash the CLI if logging setup fails

-# Apply IPv4 preference early, before any HTTP clients are created.
-try:
-    from hermes_cli.config import load_config as _load_config_early
-    from hermes_constants import apply_ipv4_preference as _apply_ipv4
-    _early_cfg = _load_config_early()
-    _net = _early_cfg.get("network", {})
-    if isinstance(_net, dict) and _net.get("force_ipv4"):
-        _apply_ipv4(force=True)
-    del _early_cfg, _net
-except Exception:
-    pass  # best-effort — don't crash if config isn't available yet
-
 import logging
 import time as _time
 from datetime import datetime
@@ -540,113 +528,6 @@ def _resolve_last_cli_session() -> Optional[str]:
    return None


-def _probe_container(cmd: list, backend: str, via_sudo: bool = False):
-    """Run a container inspect probe, returning the CompletedProcess.
-
-    Catches TimeoutExpired specifically for a human-readable message;
-    all other exceptions propagate naturally.
-    """
-    try:
-        return subprocess.run(cmd, capture_output=True, text=True, timeout=15)
-    except subprocess.TimeoutExpired:
-        label = f"sudo {backend}" if via_sudo else backend
-        print(
-            f"Error: timed out waiting for {label} to respond.\n"
-            f"The {backend} daemon may be unresponsive or starting up.",
-            file=sys.stderr,
-        )
-        sys.exit(1)
-
-
-def _exec_in_container(container_info: dict, cli_args: list):
-    """Replace the current process with a command inside the managed container.
-
-    Probes whether sudo is needed (rootful containers), then os.execvp
-    into the container. On success the Python process is replaced entirely
-    and the container's exit code becomes the process exit code (OS semantics).
-    On failure, OSError propagates naturally.
-
-    Args:
-        container_info: dict with backend, container_name, exec_user, hermes_bin
-        cli_args: the original CLI arguments (everything after 'hermes')
-    """
-    import shutil
-
-    backend = container_info["backend"]
-    container_name = container_info["container_name"]
-    exec_user = container_info["exec_user"]
-    hermes_bin = container_info["hermes_bin"]
-
-    runtime = shutil.which(backend)
-    if not runtime:
-        print(f"Error: {backend} not found on PATH. Cannot route to container.",
-              file=sys.stderr)
-        sys.exit(1)
-
-    # Rootful containers (NixOS systemd service) are invisible to unprivileged
-    # users — Podman uses per-user namespaces, Docker needs group access.
-    # Probe whether the runtime can see the container; if not, try via sudo.
-    sudo_path = None
-    probe = _probe_container(
-        [runtime, "inspect", "--format", "ok", container_name], backend,
-    )
-    if probe.returncode != 0:
-        sudo_path = shutil.which("sudo")
-        if sudo_path:
-            probe2 = _probe_container(
-                [sudo_path, "-n", runtime, "inspect", "--format", "ok", container_name],
-                backend, via_sudo=True,
-            )
-            if probe2.returncode != 0:
-                print(
-                    f"Error: container '{container_name}' not found via {backend}.\n"
-                    f"\n"
-                    f"The container is likely running as root. Your user cannot see it\n"
-                    f"because {backend} uses per-user namespaces. Grant passwordless\n"
-                    f"sudo for {backend} — the -n (non-interactive) flag is required\n"
-                    f"because a password prompt would hang or break piped commands.\n"
-                    f"\n"
-                    f"On NixOS:\n"
-                    f"\n"
-                    f'  security.sudo.extraRules = [{{\n'
-                    f'    users = [ "{os.getenv("USER", "your-user")}" ];\n'
-                    f'    commands = [{{ command = "{runtime}"; options = [ "NOPASSWD" ]; }}];\n'
-                    f'  }}];\n'
-                    f"\n"
-                    f"Or run: sudo hermes {' '.join(cli_args)}",
-                    file=sys.stderr,
-                )
-                sys.exit(1)
-        else:
-            print(
-                f"Error: container '{container_name}' not found via {backend}.\n"
-                f"The container may be running under root. Try: sudo hermes {' '.join(cli_args)}",
-                file=sys.stderr,
-            )
-            sys.exit(1)
-
-    is_tty = sys.stdin.isatty()
-    tty_flags = ["-it"] if is_tty else ["-i"]
-
-    env_flags = []
-    for var in ("TERM", "COLORTERM", "LANG", "LC_ALL"):
-        val = os.environ.get(var)
-        if val:
-            env_flags.extend(["-e", f"{var}={val}"])
-
-    cmd_prefix = [sudo_path, "-n", runtime] if sudo_path else [runtime]
-    exec_cmd = (
-        cmd_prefix + ["exec"]
-        + tty_flags
-        + ["-u", exec_user]
-        + env_flags
-        + [container_name, hermes_bin]
-        + cli_args
-    )
-
-    os.execvp(exec_cmd[0], exec_cmd)
-
-
 def _resolve_session_by_name_or_id(name_or_id: str) -> Optional[str]:
    """Resolve a session name (title) or ID to a session ID.

@@ -1053,7 +934,6 @@ def select_provider_and_model(args=None):
        "kilocode": "Kilo Code",
        "alibaba": "Alibaba Cloud (DashScope)",
        "huggingface": "Hugging Face",
-        "xiaomi": "Xiaomi MiMo",
        "custom": "Custom endpoint",
    }
    active_label = provider_labels.get(active, active) if active else "none"
@@ -1086,7 +966,6 @@ def select_provider_and_model(args=None):
        ("opencode-go", "OpenCode Go (open models, $10/month subscription)"),
        ("ai-gateway", "AI Gateway (Vercel — 200+ models, pay-per-use)"),
        ("alibaba", "Alibaba Cloud / DashScope Coding (Qwen + multi-provider)"),
-        ("xiaomi", "Xiaomi MiMo (MiMo-V2 models — pro, omni, flash)"),
    ]

    def _named_custom_provider_map(cfg) -> dict[str, dict[str, str]]:
@@ -1198,45 +1077,9 @@ def select_provider_and_model(args=None):
        _model_flow_anthropic(config, current_model)
    elif selected_provider == "kimi-coding":
        _model_flow_kimi(config, current_model)
-    elif selected_provider in ("gemini", "zai", "minimax", "minimax-cn", "kilocode", "opencode-zen", "opencode-go", "ai-gateway", "alibaba", "huggingface", "xiaomi"):
+    elif selected_provider in ("gemini", "zai", "minimax", "minimax-cn", "kilocode", "opencode-zen", "opencode-go", "ai-gateway", "alibaba", "huggingface"):
        _model_flow_api_key_provider(config, selected_provider, current_model)

-    # ── Post-switch cleanup: clear stale OPENAI_BASE_URL ──────────────
-    # When the user switches to a named provider (anything except "custom"),
-    # a leftover OPENAI_BASE_URL in ~/.hermes/.env can poison auxiliary
-    # clients that use provider:auto. Clear it proactively.  (#5161)
-    if selected_provider not in ("custom", "cancel", "remove-custom") \
-            and not selected_provider.startswith("custom:"):
-        _clear_stale_openai_base_url()
-
-
-def _clear_stale_openai_base_url():
-    """Remove OPENAI_BASE_URL from ~/.hermes/.env if the active provider is not 'custom'.
-
-    After a provider switch, a leftover OPENAI_BASE_URL causes auxiliary
-    clients (compression, vision, delegation) with provider:auto to route
-    requests to the old custom endpoint instead of the newly selected
-    provider.  See issue #5161.
-    """
-    from hermes_cli.config import get_env_value, save_env_value, load_config
-
-    cfg = load_config()
-    model_cfg = cfg.get("model", {})
-    if isinstance(model_cfg, dict):
-        provider = (model_cfg.get("provider") or "").strip().lower()
-    else:
-        provider = ""
-
-    if provider == "custom" or not provider:
-        return  # custom provider legitimately uses OPENAI_BASE_URL
-
-    stale_url = get_env_value("OPENAI_BASE_URL")
-    if stale_url:
-        save_env_value("OPENAI_BASE_URL", "")
-        print(f"Cleared stale OPENAI_BASE_URL from .env (was: {stale_url[:40]}...)"
-              if len(stale_url) > 40
-              else f"Cleared stale OPENAI_BASE_URL from .env (was: {stale_url})")
-

 def _prompt_provider_choice(choices, *, default=0):
    """Show provider selection menu with curses arrow-key navigation.
@@ -2668,8 +2511,13 @@ def _model_flow_anthropic(config, current_model=""):
    from hermes_cli.models import _PROVIDER_MODELS

    # Check ALL credential sources
-    from hermes_cli.auth import get_anthropic_key
-    existing_key = get_anthropic_key()
+    existing_key = (
+        get_env_value("ANTHROPIC_TOKEN")
+        or os.getenv("ANTHROPIC_TOKEN", "")
+        or get_env_value("ANTHROPIC_API_KEY")
+        or os.getenv("ANTHROPIC_API_KEY", "")
+        or os.getenv("CLAUDE_CODE_OAUTH_TOKEN", "")
+    )
    cc_available = False
    try:
        from agent.anthropic_adapter import read_claude_code_credentials, is_claude_code_token_valid
@@ -2830,18 +2678,6 @@ def cmd_config(args):
    config_command(args)


-def cmd_backup(args):
-    """Back up Hermes home directory to a zip file."""
-    from hermes_cli.backup import run_backup
-    run_backup(args)
-
-
-def cmd_import(args):
-    """Restore a Hermes backup from a zip file."""
-    from hermes_cli.backup import run_import
-    run_import(args)
-
-
 def cmd_version(args):
    """Show version."""
    print(f"Hermes Agent v{__version__} ({__release_date__})")
@@ -3054,7 +2890,23 @@ def _update_via_zip(args):
                check=True,
            )
        _install_python_dependencies_with_optional_fallback(pip_cmd)
-    
+
+    # Build web UI frontend (optional — requires npm)
+    web_dir = PROJECT_ROOT / "web"
+    if (web_dir / "package.json").exists():
+        import shutil as _shutil
+        if _shutil.which("npm"):
+            print("→ Building web UI...")
+            r1 = subprocess.run(["npm", "install", "--silent"], cwd=web_dir, capture_output=True)
+            if r1.returncode == 0:
+                r2 = subprocess.run(["npm", "run", "build"], cwd=web_dir, capture_output=True)
+                if r2.returncode == 0:
+                    print("  ✓ Web UI built")
+                else:
+                    print("  ⚠ Web UI build failed (hermes web will not be available)")
+            else:
+                print("  ⚠ Web UI npm install failed (hermes web will not be available)")
+
    # Sync skills
    try:
        from tools.skills_sync import sync_skills
@@ -3801,7 +3653,23 @@ def cmd_update(args):
            if shutil.which("npm"):
                print("→ Updating Node.js dependencies...")
                subprocess.run(["npm", "install", "--silent"], cwd=PROJECT_ROOT, check=False)
-        
+
+        # Build web UI frontend (optional — requires npm)
+        web_dir = PROJECT_ROOT / "web"
+        if (web_dir / "package.json").exists():
+            import shutil as _shutil_web
+            if _shutil_web.which("npm"):
+                print("→ Building web UI...")
+                r1 = subprocess.run(["npm", "install", "--silent"], cwd=web_dir, capture_output=True)
+                if r1.returncode == 0:
+                    r2 = subprocess.run(["npm", "run", "build"], cwd=web_dir, capture_output=True)
+                    if r2.returncode == 0:
+                        print("  ✓ Web UI built")
+                    else:
+                        print("  ⚠ Web UI build failed (hermes web will not be available)")
+                else:
+                    print("  ⚠ Web UI npm install failed (hermes web will not be available)")
+
        print()
        print("✓ Code updated!")
        
@@ -4007,7 +3875,7 @@ def cmd_update(args):
            # Exclude PIDs that belong to just-restarted services so we don't
            # immediately kill the process that systemd/launchd just spawned.
            service_pids = _get_service_pids()
-            manual_pids = find_gateway_pids(exclude_pids=service_pids, all_profiles=True)
+            manual_pids = find_gateway_pids(exclude_pids=service_pids)
            for pid in manual_pids:
                try:
                    os.kill(pid, _signal.SIGTERM)
@@ -4063,7 +3931,7 @@ def _coalesce_session_name_args(argv: list) -> list:
        "chat", "model", "gateway", "setup", "whatsapp", "login", "logout", "auth",
        "status", "cron", "doctor", "config", "pairing", "skills", "tools",
        "mcp", "sessions", "insights", "version", "update", "uninstall",
-        "profile",
+        "profile", "web",
    }
    _SESSION_FLAGS = {"-c", "--continue", "-r", "--resume"}

@@ -4335,6 +4203,50 @@ def cmd_profile(args):
            sys.exit(1)


+def cmd_web(args):
+    """Start the web UI server."""
+    try:
+        import fastapi  # noqa: F401
+        import uvicorn  # noqa: F401
+    except ImportError:
+        print("Web UI dependencies not installed.")
+        print("Install them with:  pip install hermes-agent[web]")
+        sys.exit(1)
+
+    web_dist = PROJECT_ROOT / "hermes_cli" / "web_dist"
+    web_src = PROJECT_ROOT / "web"
+    if not web_dist.exists() and (web_src / "package.json").exists():
+        import shutil
+        npm = shutil.which("npm")
+        if npm:
+            import subprocess
+            print("→ Web UI not built yet — building now...")
+            r1 = subprocess.run([npm, "install", "--silent"], cwd=web_src, capture_output=True)
+            if r1.returncode == 0:
+                r2 = subprocess.run([npm, "run", "build"], cwd=web_src, capture_output=True)
+                if r2.returncode == 0:
+                    print("  ✓ Web UI built")
+                else:
+                    print("  ✗ Web UI build failed")
+                    print("  Run manually:  cd web && npm install && npm run build")
+                    sys.exit(1)
+            else:
+                print("  ✗ npm install failed")
+                print("  Run manually:  cd web && npm install && npm run build")
+                sys.exit(1)
+        else:
+            print("Web UI frontend not built and npm is not available.")
+            print("Install Node.js, then run:  cd web && npm install && npm run build")
+            sys.exit(1)
+
+    from hermes_cli.web_server import start_server
+    start_server(
+        host=args.host,
+        port=args.port,
+        open_browser=not args.no_open,
+    )
+
+
 def cmd_completion(args):
    """Print shell completion script."""
    from hermes_cli.profiles import generate_bash_completion, generate_zsh_completion
@@ -4362,7 +4274,6 @@ def cmd_logs(args):
        level=getattr(args, "level", None),
        session=getattr(args, "session", None),
        since=getattr(args, "since", None),
-        component=getattr(args, "component", None),
    )


@@ -4486,7 +4397,7 @@ For more help on a command:
    )
    chat_parser.add_argument(
        "--provider",
-        choices=["auto", "openrouter", "nous", "openai-codex", "copilot-acp", "copilot", "anthropic", "gemini", "huggingface", "zai", "kimi-coding", "minimax", "minimax-cn", "kilocode", "xiaomi"],
+        choices=["auto", "openrouter", "nous", "openai-codex", "copilot-acp", "copilot", "anthropic", "gemini", "huggingface", "zai", "kimi-coding", "minimax", "minimax-cn", "kilocode"],
        default=None,
        help="Inference provider (default: auto)"
    )
@@ -4928,43 +4839,7 @@ For more help on a command:
        help="Show redacted API key prefixes (first/last 4 chars) instead of just set/not set"
    )
    dump_parser.set_defaults(func=cmd_dump)
-
-    # =========================================================================
-    # backup command
-    # =========================================================================
-    backup_parser = subparsers.add_parser(
-        "backup",
-        help="Back up Hermes home directory to a zip file",
-        description="Create a zip archive of your entire Hermes configuration, "
-                    "skills, sessions, and data (excludes the hermes-agent codebase)"
-    )
-    backup_parser.add_argument(
-        "-o", "--output",
-        help="Output path for the zip file (default: ~/hermes-backup-<timestamp>.zip)"
-    )
-    backup_parser.set_defaults(func=cmd_backup)
-
-    # =========================================================================
-    # import command
-    # =========================================================================
-    import_parser = subparsers.add_parser(
-        "import",
-        help="Restore a Hermes backup from a zip file",
-        description="Extract a previously created Hermes backup into your "
-                    "Hermes home directory, restoring configuration, skills, "
-                    "sessions, and data"
-    )
-    import_parser.add_argument(
-        "zipfile",
-        help="Path to the backup zip file"
-    )
-    import_parser.add_argument(
-        "--force", "-f",
-        action="store_true",
-        help="Overwrite existing files without confirmation"
-    )
-    import_parser.set_defaults(func=cmd_import)
-
+    
    # =========================================================================
    # config command
    # =========================================================================
@@ -5314,8 +5189,6 @@ For more help on a command:
    mcp_add_p.add_argument("--command", help="Stdio command (e.g. npx)")
    mcp_add_p.add_argument("--args", nargs="*", default=[], help="Arguments for stdio command")
    mcp_add_p.add_argument("--auth", choices=["oauth", "header"], help="Auth method")
-    mcp_add_p.add_argument("--preset", help="Known MCP preset name")
-    mcp_add_p.add_argument("--env", nargs="*", default=[], help="Environment variables for stdio servers (KEY=VALUE)")

    mcp_rm_p = mcp_sub.add_parser("remove", aliases=["rm"], help="Remove an MCP server")
    mcp_rm_p.add_argument("name", help="Server name to remove")
@@ -5578,8 +5451,7 @@ For more help on a command:
    claw_migrate = claw_subparsers.add_parser(
        "migrate",
        help="Migrate from OpenClaw to Hermes",
-        description="Import settings, memories, skills, and API keys from an OpenClaw installation. "
-                    "Always shows a preview before making changes."
+        description="Import settings, memories, skills, and API keys from an OpenClaw installation"
    )
    claw_migrate.add_argument(
        "--source",
@@ -5588,7 +5460,7 @@ For more help on a command:
    claw_migrate.add_argument(
        "--dry-run",
        action="store_true",
-        help="Preview only — stop after showing what would be migrated"
+        help="Preview what would be migrated without making changes"
    )
    claw_migrate.add_argument(
        "--preset",
@@ -5782,6 +5654,19 @@ For more help on a command:
    )
    completion_parser.set_defaults(func=cmd_completion)

+    # =========================================================================
+    # web command
+    # =========================================================================
+    web_parser = subparsers.add_parser(
+        "web",
+        help="Start the web UI dashboard",
+        description="Launch the Hermes Agent web dashboard for managing config, API keys, and sessions",
+    )
+    web_parser.add_argument("--port", type=int, default=9119, help="Port (default 9119)")
+    web_parser.add_argument("--host", default="127.0.0.1", help="Host (default 127.0.0.1)")
+    web_parser.add_argument("--no-open", action="store_true", help="Don't open browser automatically")
+    web_parser.set_defaults(func=cmd_web)
+
    # =========================================================================
    # logs command
    # =========================================================================
@@ -5798,7 +5683,6 @@ Examples:
    hermes logs gateway -n 100     Show last 100 lines of gateway.log
    hermes logs --level WARNING    Only show WARNING and above
    hermes logs --session abc123   Filter by session ID
-    hermes logs --component tools  Only show tool-related lines
    hermes logs --since 1h         Lines from the last hour
    hermes logs --since 30m -f     Follow, starting from 30 min ago
    hermes logs list               List available log files with sizes
@@ -5828,10 +5712,6 @@ Examples:
        "--since", metavar="TIME",
        help="Show lines since TIME ago (e.g. 1h, 30m, 2d)",
    )
-    logs_parser.add_argument(
-        "--component", metavar="NAME",
-        help="Filter by component: gateway, agent, tools, cli, cron",
-    )
    logs_parser.set_defaults(func=cmd_logs)

    # =========================================================================
@@ -5840,22 +5720,9 @@ Examples:
    # Pre-process argv so unquoted multi-word session names after -c / -r
    # are merged into a single token before argparse sees them.
    # e.g. ``hermes -c Pokemon Agent Dev`` → ``hermes -c 'Pokemon Agent Dev'``
-    # ── Container-aware routing ────────────────────────────────────────
-    # When NixOS container mode is active, route ALL subcommands into
-    # the managed container.  This MUST run before parse_args() so that
-    # --help, unrecognised flags, and every subcommand are forwarded
-    # transparently instead of being intercepted by argparse on the host.
-    from hermes_cli.config import get_container_exec_info
-    container_info = get_container_exec_info()
-    if container_info:
-        _exec_in_container(container_info, sys.argv[1:])
-        # Unreachable: os.execvp never returns on success (process is replaced)
-        # and raises OSError on failure (which propagates as a traceback).
-        sys.exit(1)
-
    _processed_argv = _coalesce_session_name_args(sys.argv[1:])
    args = parser.parse_args(_processed_argv)
-
+    
    # Handle --version flag
    if args.version:
        cmd_version(args)
@@ -9,6 +9,7 @@ configuration in ~/.hermes/config.yaml under the ``mcp_servers`` key.
 """

 import asyncio
+import getpass
 import logging
 import os
 import re
@@ -27,11 +28,6 @@ from hermes_constants import display_hermes_home

 logger = logging.getLogger(__name__)

-_ENV_VAR_NAME_RE = re.compile(r"^[A-Za-z_][A-Za-z0-9_]*$")
-
-
-_MCP_PRESETS: Dict[str, Dict[str, Any]] = {}
-

 # ─── UI Helpers ───────────────────────────────────────────────────────────────

@@ -61,8 +57,19 @@ def _confirm(question: str, default: bool = True) -> bool:


 def _prompt(question: str, *, password: bool = False, default: str = "") -> str:
-    from hermes_cli.cli_output import prompt as _shared_prompt
-    return _shared_prompt(question, default=default, password=password)
+    display = f"  {question}"
+    if default:
+        display += f" [{default}]"
+    display += ": "
+    try:
+        if password:
+            value = getpass.getpass(color(display, Colors.YELLOW))
+        else:
+            value = input(color(display, Colors.YELLOW))
+        return value.strip() or default
+    except (KeyboardInterrupt, EOFError):
+        print()
+        return default


 # ─── Config Helpers ───────────────────────────────────────────────────────────
@@ -102,59 +109,6 @@ def _env_key_for_server(name: str) -> str:
    return f"MCP_{name.upper().replace('-', '_')}_API_KEY"


-def _parse_env_assignments(raw_env: Optional[List[str]]) -> Dict[str, str]:
-    """Parse ``KEY=VALUE`` strings from CLI args into an env dict."""
-    parsed: Dict[str, str] = {}
-    for item in raw_env or []:
-        text = str(item or "").strip()
-        if not text:
-            continue
-        if "=" not in text:
-            raise ValueError(f"Invalid --env value '{text}' (expected KEY=VALUE)")
-        key, value = text.split("=", 1)
-        key = key.strip()
-        if not key:
-            raise ValueError(f"Invalid --env value '{text}' (missing variable name)")
-        if not _ENV_VAR_NAME_RE.match(key):
-            raise ValueError(f"Invalid --env variable name '{key}'")
-        parsed[key] = value
-    return parsed
-
-
-def _apply_mcp_preset(
-    name: str,
-    *,
-    preset_name: Optional[str],
-    url: Optional[str],
-    command: Optional[str],
-    cmd_args: List[str],
-    server_config: Dict[str, Any],
-) -> tuple[Optional[str], Optional[str], List[str], bool]:
-    """Apply a known MCP preset when transport details were omitted."""
-    if not preset_name:
-        return url, command, cmd_args, False
-
-    preset = _MCP_PRESETS.get(preset_name)
-    if not preset:
-        raise ValueError(f"Unknown MCP preset: {preset_name}")
-
-    if url or command:
-        return url, command, cmd_args, False
-
-    url = preset.get("url")
-    command = preset.get("command")
-    cmd_args = list(preset.get("args") or [])
-
-    if url:
-        server_config["url"] = url
-    if command:
-        server_config["command"] = command
-    if cmd_args:
-        server_config["args"] = cmd_args
-
-    return url, command, cmd_args, True
-
-
 # ─── Discovery (temporary connect) ───────────────────────────────────────────

 def _probe_single_server(
@@ -223,35 +177,13 @@ def cmd_mcp_add(args):
    command = getattr(args, "command", None)
    cmd_args = getattr(args, "args", None) or []
    auth_type = getattr(args, "auth", None)
-    preset_name = getattr(args, "preset", None)
-    raw_env = getattr(args, "env", None)
-
-    server_config: Dict[str, Any] = {}
-    try:
-        explicit_env = _parse_env_assignments(raw_env)
-        url, command, cmd_args, _preset_applied = _apply_mcp_preset(
-            name,
-            preset_name=preset_name,
-            url=url,
-            command=command,
-            cmd_args=list(cmd_args),
-            server_config=server_config,
-        )
-    except ValueError as exc:
-        _error(str(exc))
-        return
-
-    if url and explicit_env:
-        _error("--env is only supported for stdio MCP servers (--command or stdio presets)")
-        return

    # Validate transport
    if not url and not command:
-        _error("Must specify --url <endpoint>, --command <cmd>, or --preset <name>")
+        _error("Must specify --url <endpoint> or --command <cmd>")
        _info("Examples:")
        _info('  hermes mcp add ink --url "https://mcp.ml.ink/mcp"')
        _info('  hermes mcp add github --command npx --args @modelcontextprotocol/server-github')
-        _info('  hermes mcp add myserver --preset mypreset')
        return

    # Check if server already exists
@@ -262,15 +194,13 @@ def cmd_mcp_add(args):
            return

    # Build initial config
+    server_config: Dict[str, Any] = {}
    if url:
        server_config["url"] = url
    else:
        server_config["command"] = command
        if cmd_args:
            server_config["args"] = cmd_args
-        if explicit_env:
-            server_config["env"] = explicit_env
-

    # ── Authentication ────────────────────────────────────────────────

@@ -708,7 +638,6 @@ def mcp_command(args):
        _info("hermes mcp serve                              Run as MCP server")
        _info("hermes mcp add <name> --url <endpoint>        Add an MCP server")
        _info("hermes mcp add <name> --command <cmd>         Add a stdio server")
-        _info("hermes mcp add <name> --preset <preset>       Add from a known preset")
        _info("hermes mcp remove <name>                      Remove a server")
        _info("hermes mcp list                               List servers")
        _info("hermes mcp test <name>                        Test connection")
@@ -25,13 +25,85 @@ def _curses_select(title: str, items: list[tuple[str, str]], default: int = 0) -
    items: list of (label, description) tuples.
    Returns selected index, or default on escape/quit.
    """
-    from hermes_cli.curses_ui import curses_radiolist
-    # Format (label, desc) tuples into display strings
-    display_items = [
-        f"{label}  {desc}" if desc else label
-        for label, desc in items
-    ]
-    return curses_radiolist(title, display_items, selected=default, cancel_returns=default)
+    try:
+        import curses
+        result = [default]
+
+        def _menu(stdscr):
+            curses.curs_set(0)
+            if curses.has_colors():
+                curses.start_color()
+                curses.use_default_colors()
+                curses.init_pair(1, curses.COLOR_GREEN, -1)
+                curses.init_pair(2, curses.COLOR_YELLOW, -1)
+                curses.init_pair(3, curses.COLOR_CYAN, -1)
+            cursor = default
+
+            while True:
+                stdscr.clear()
+                max_y, max_x = stdscr.getmaxyx()
+
+                # Title
+                try:
+                    stdscr.addnstr(0, 0, title, max_x - 1,
+                                   curses.A_BOLD | (curses.color_pair(2) if curses.has_colors() else 0))
+                    stdscr.addnstr(1, 0, "  ↑↓ navigate  ⏎ select  q quit", max_x - 1,
+                                   curses.color_pair(3) if curses.has_colors() else curses.A_DIM)
+                except curses.error:
+                    pass
+
+                for i, (label, desc) in enumerate(items):
+                    y = i + 3
+                    if y >= max_y - 1:
+                        break
+                    arrow = "→" if i == cursor else " "
+                    line = f" {arrow}  {label}"
+                    if desc:
+                        line += f"  {desc}"
+
+                    attr = curses.A_NORMAL
+                    if i == cursor:
+                        attr = curses.A_BOLD
+                        if curses.has_colors():
+                            attr |= curses.color_pair(1)
+                    try:
+                        stdscr.addnstr(y, 0, line[:max_x - 1], max_x - 1, attr)
+                    except curses.error:
+                        pass
+
+                stdscr.refresh()
+                key = stdscr.getch()
+
+                if key in (curses.KEY_UP, ord('k')):
+                    cursor = (cursor - 1) % len(items)
+                elif key in (curses.KEY_DOWN, ord('j')):
+                    cursor = (cursor + 1) % len(items)
+                elif key in (curses.KEY_ENTER, 10, 13):
+                    result[0] = cursor
+                    return
+                elif key in (27, ord('q')):
+                    return
+
+        curses.wrapper(_menu)
+        return result[0]
+
+    except Exception:
+        # Fallback: numbered input
+        print(f"\n  {title}\n")
+        for i, (label, desc) in enumerate(items):
+            marker = "→" if i == default else " "
+            d = f"  {desc}" if desc else ""
+            print(f"  {marker} {i + 1}. {label}{d}")
+        while True:
+            try:
+                val = input(f"\n  Select [1-{len(items)}] ({default + 1}): ")
+                if not val:
+                    return default
+                idx = int(val) - 1
+                if 0 <= idx < len(items):
+                    return idx
+            except (ValueError, EOFError):
+                return default


 def _prompt(label: str, default: str | None = None, secret: bool = False) -> str:
@@ -74,13 +74,13 @@ _DOT_TO_HYPHEN_PROVIDERS: frozenset[str] = frozenset({
 _STRIP_VENDOR_ONLY_PROVIDERS: frozenset[str] = frozenset({
    "copilot",
    "copilot-acp",
-    "openai-codex",
 })

 # Providers whose native naming is authoritative -- pass through unchanged.
 _AUTHORITATIVE_NATIVE_PROVIDERS: frozenset[str] = frozenset({
    "gemini",
    "huggingface",
+    "openai-codex",
 })

 # Direct providers that accept bare native names but should repair a matching
@@ -92,7 +92,6 @@ _MATCHING_PREFIX_STRIP_PROVIDERS: frozenset[str] = frozenset({
    "minimax-cn",
    "alibaba",
    "qwen-oauth",
-    "xiaomi",
    "custom",
 })

@@ -360,11 +359,7 @@ def normalize_model_for_provider(model_input: str, target_provider: str) -> str:

    # --- Copilot: strip matching provider prefix, keep dots ---
    if provider in _STRIP_VENDOR_ONLY_PROVIDERS:
-        stripped = _strip_matching_provider_prefix(name, provider)
-        if stripped == name and name.startswith("openai/"):
-            # openai-codex maps openai/gpt-5.4 -> gpt-5.4
-            return name.split("/", 1)[1]
-        return stripped
+        return _strip_matching_provider_prefix(name, provider)

    # --- DeepSeek: map to one of two canonical names ---
    if provider == "deepseek":
@@ -839,11 +839,8 @@ def list_authenticated_providers(
                    if any(os.environ.get(ev) for ev in pcfg.api_key_env_vars):
                        has_creds = True
                        break
-        # Check auth store and credential pool for non-env-var credentials.
-        # This applies to OAuth providers AND api_key providers that also
-        # support OAuth (e.g. anthropic supports both API key and Claude Code
-        # OAuth via external credential files).
-        if not has_creds:
+        if not has_creds and overlay.auth_type in ("oauth_device_code", "oauth_external", "external_process"):
+            # These use auth stores, not env vars — check for auth.json entries
            try:
                from hermes_cli.auth import _load_auth_store
                store = _load_auth_store()
@@ -856,38 +853,6 @@ def list_authenticated_providers(
                    has_creds = True
            except Exception as exc:
                logger.debug("Auth store check failed for %s: %s", pid, exc)
-        # Fallback: check the credential pool with full auto-seeding.
-        # This catches credentials that exist in external stores (e.g.
-        # Codex CLI ~/.codex/auth.json) which _seed_from_singletons()
-        # imports on demand but aren't in the raw auth.json yet.
-        if not has_creds:
-            try:
-                from agent.credential_pool import load_pool
-                pool = load_pool(hermes_slug)
-                if pool.has_credentials():
-                    has_creds = True
-            except Exception as exc:
-                logger.debug("Credential pool check failed for %s: %s", hermes_slug, exc)
-        # Fallback: check external credential files directly.
-        # The credential pool gates anthropic behind
-        # is_provider_explicitly_configured() to prevent auxiliary tasks
-        # from silently consuming Claude Code tokens (PR #4210).
-        # But the /model picker is discovery-oriented — we WANT to show
-        # providers the user can switch to, even if they aren't currently
-        # configured.
-        if not has_creds and hermes_slug == "anthropic":
-            try:
-                from agent.anthropic_adapter import (
-                    read_claude_code_credentials,
-                    read_hermes_oauth_credentials,
-                )
-                hermes_creds = read_hermes_oauth_credentials()
-                cc_creds = read_claude_code_credentials()
-                if (hermes_creds and hermes_creds.get("accessToken")) or \
-                   (cc_creds and cc_creds.get("accessToken")):
-                    has_creds = True
-            except Exception as exc:
-                logger.debug("Anthropic external creds check failed: %s", exc)
        if not has_creds:
            continue

@@ -56,18 +56,6 @@ OPENROUTER_MODELS: list[tuple[str, str]] = [

 _openrouter_catalog_cache: list[tuple[str, str]] | None = None

-
-def _codex_curated_models() -> list[str]:
-    """Derive the openai-codex curated list from codex_models.py.
-
-    Single source of truth: DEFAULT_CODEX_MODELS + forward-compat synthesis.
-    This keeps the gateway /model picker in sync with the CLI `hermes model`
-    flow without maintaining a separate static list.
-    """
-    from hermes_cli.codex_models import DEFAULT_CODEX_MODELS, _add_forward_compat_models
-    return _add_forward_compat_models(list(DEFAULT_CODEX_MODELS))
-
-
 _PROVIDER_MODELS: dict[str, list[str]] = {
    "nous": [
        "anthropic/claude-opus-4.6",
@@ -98,7 +86,12 @@ _PROVIDER_MODELS: dict[str, list[str]] = {
        "openai/gpt-5.4-pro",
        "openai/gpt-5.4-nano",
    ],
-    "openai-codex": _codex_curated_models(),
+    "openai-codex": [
+        "gpt-5.3-codex",
+        "gpt-5.2-codex",
+        "gpt-5.1-codex-mini",
+        "gpt-5.1-codex-max",
+    ],
    "copilot-acp": [
        "copilot-acp",
    ],
@@ -164,16 +157,22 @@ _PROVIDER_MODELS: dict[str, list[str]] = {
        "kimi-k2-0905-preview",
    ],
    "minimax": [
-        "MiniMax-M2.7",
+        "MiniMax-M1",
+        "MiniMax-M1-40k",
+        "MiniMax-M1-80k",
+        "MiniMax-M1-128k",
+        "MiniMax-M1-256k",
        "MiniMax-M2.5",
-        "MiniMax-M2.1",
-        "MiniMax-M2",
+        "MiniMax-M2.7",
    ],
    "minimax-cn": [
-        "MiniMax-M2.7",
+        "MiniMax-M1",
+        "MiniMax-M1-40k",
+        "MiniMax-M1-80k",
+        "MiniMax-M1-128k",
+        "MiniMax-M1-256k",
        "MiniMax-M2.5",
-        "MiniMax-M2.1",
-        "MiniMax-M2",
+        "MiniMax-M2.7",
    ],
    "anthropic": [
        "claude-opus-4-6",
@@ -188,11 +187,6 @@ _PROVIDER_MODELS: dict[str, list[str]] = {
        "deepseek-chat",
        "deepseek-reasoner",
    ],
-    "xiaomi": [
-        "mimo-v2-pro",
-        "mimo-v2-omni",
-        "mimo-v2-flash",
-    ],
    "opencode-zen": [
        "gpt-5.4-pro",
        "gpt-5.4",
@@ -498,7 +492,6 @@ _PROVIDER_LABELS = {
    "alibaba": "Alibaba Cloud (DashScope)",
    "qwen-oauth": "Qwen OAuth (Portal)",
    "huggingface": "Hugging Face",
-    "xiaomi": "Xiaomi MiMo",
    "custom": "Custom endpoint",
 }

@@ -541,8 +534,6 @@ _PROVIDER_ALIASES = {
    "hf": "huggingface",
    "hugging-face": "huggingface",
    "huggingface-hub": "huggingface",
-    "mimo": "xiaomi",
-    "xiaomi-mimo": "xiaomi",
 }


@@ -827,7 +818,7 @@ def list_available_providers() -> list[dict[str, str]]:
        "openrouter", "nous", "openai-codex", "copilot", "copilot-acp",
        "gemini", "huggingface",
        "zai", "kimi-coding", "minimax", "minimax-cn", "kilocode", "anthropic", "alibaba",
-        "qwen-oauth", "xiaomi",
+        "qwen-oauth",
        "opencode-zen", "opencode-go",
        "ai-gateway", "deepseek", "custom",
    ]
@@ -1809,35 +1800,6 @@ def validate_requested_model(
            "message": message,
        }

-    # OpenAI Codex has its own catalog path; /v1/models probing is not the right validation path.
-    if normalized == "openai-codex":
-        try:
-            codex_models = provider_model_ids("openai-codex")
-        except Exception:
-            codex_models = []
-        if codex_models:
-            if requested_for_lookup in set(codex_models):
-                return {
-                    "accepted": True,
-                    "persist": True,
-                    "recognized": True,
-                    "message": None,
-                }
-            suggestions = get_close_matches(requested_for_lookup, codex_models, n=3, cutoff=0.5)
-            suggestion_text = ""
-            if suggestions:
-                suggestion_text = "\n  Similar models: " + ", ".join(f"`{s}`" for s in suggestions)
-            return {
-                "accepted": True,
-                "persist": True,
-                "recognized": False,
-                "message": (
-                    f"Note: `{requested}` was not found in the OpenAI Codex model listing. "
-                    f"It may still work if your account has access to it."
-                    f"{suggestion_text}"
-                ),
-            }
-
    # Probe the live API to check if the model actually exists
    api_models = fetch_api_models(api_key, base_url)

@@ -143,7 +143,6 @@ def _tts_label(current_provider: str) -> str:
        "openai": "OpenAI TTS",
        "elevenlabs": "ElevenLabs",
        "edge": "Edge TTS",
-        "mistral": "Mistral Voxtral TTS",
        "neutts": "NeuTTS",
    }
    return mapping.get(current_provider or "edge", current_provider or "Edge TTS")
@@ -310,7 +309,6 @@ def get_nous_subscription_features(
        tts_current_provider in {"edge", "neutts"}
        or (tts_current_provider == "openai" and (managed_tts_available or direct_openai_tts))
        or (tts_current_provider == "elevenlabs" and direct_elevenlabs)
-        or (tts_current_provider == "mistral" and bool(get_env_value("MISTRAL_API_KEY")))
    )
    tts_active = bool(tts_tool_enabled and tts_available)

@@ -1,46 +0,0 @@
-"""
-Shared platform registry for Hermes Agent.
-
-Single source of truth for platform metadata consumed by both
-skills_config (label display) and tools_config (default toolset
-resolution).  Import ``PLATFORMS`` from here instead of maintaining
-duplicate dicts in each module.
-"""
-
-from collections import OrderedDict
-from typing import NamedTuple
-
-
-class PlatformInfo(NamedTuple):
-    """Metadata for a single platform entry."""
-    label: str
-    default_toolset: str
-
-
-# Ordered so that TUI menus are deterministic.
-PLATFORMS: OrderedDict[str, PlatformInfo] = OrderedDict([
-    ("cli",            PlatformInfo(label="🖥️  CLI",            default_toolset="hermes-cli")),
-    ("telegram",       PlatformInfo(label="📱 Telegram",        default_toolset="hermes-telegram")),
-    ("discord",        PlatformInfo(label="💬 Discord",         default_toolset="hermes-discord")),
-    ("slack",          PlatformInfo(label="💼 Slack",           default_toolset="hermes-slack")),
-    ("whatsapp",       PlatformInfo(label="📱 WhatsApp",        default_toolset="hermes-whatsapp")),
-    ("signal",         PlatformInfo(label="📡 Signal",          default_toolset="hermes-signal")),
-    ("bluebubbles",    PlatformInfo(label="💙 BlueBubbles",     default_toolset="hermes-bluebubbles")),
-    ("email",          PlatformInfo(label="📧 Email",           default_toolset="hermes-email")),
-    ("homeassistant",  PlatformInfo(label="🏠 Home Assistant",  default_toolset="hermes-homeassistant")),
-    ("mattermost",     PlatformInfo(label="💬 Mattermost",      default_toolset="hermes-mattermost")),
-    ("matrix",         PlatformInfo(label="💬 Matrix",          default_toolset="hermes-matrix")),
-    ("dingtalk",       PlatformInfo(label="💬 DingTalk",        default_toolset="hermes-dingtalk")),
-    ("feishu",         PlatformInfo(label="🪽 Feishu",          default_toolset="hermes-feishu")),
-    ("wecom",          PlatformInfo(label="💬 WeCom",           default_toolset="hermes-wecom")),
-    ("wecom_callback", PlatformInfo(label="💬 WeCom Callback",  default_toolset="hermes-wecom-callback")),
-    ("weixin",         PlatformInfo(label="💬 Weixin",          default_toolset="hermes-weixin")),
-    ("webhook",        PlatformInfo(label="🔗 Webhook",         default_toolset="hermes-webhook")),
-    ("api_server",     PlatformInfo(label="🌐 API Server",      default_toolset="hermes-api-server")),
-])
-
-
-def platform_label(key: str, default: str = "") -> str:
-    """Return the display label for a platform key, or *default*."""
-    info = PLATFORMS.get(key)
-    return info.label if info is not None else default
@@ -88,11 +88,11 @@ HERMES_OVERLAYS: Dict[str, HermesOverlay] = {
        base_url_env_var="KIMI_BASE_URL",
    ),
    "minimax": HermesOverlay(
-        transport="anthropic_messages",
+        transport="openai_chat",
        base_url_env_var="MINIMAX_BASE_URL",
    ),
    "minimax-cn": HermesOverlay(
-        transport="anthropic_messages",
+        transport="openai_chat",
        base_url_env_var="MINIMAX_CN_BASE_URL",
    ),
    "deepseek": HermesOverlay(
@@ -132,10 +132,6 @@ HERMES_OVERLAYS: Dict[str, HermesOverlay] = {
        base_url_override="https://api.x.ai/v1",
        base_url_env_var="XAI_BASE_URL",
    ),
-    "xiaomi": HermesOverlay(
-        transport="openai_chat",
-        base_url_env_var="XIAOMI_BASE_URL",
-    ),
 }


@@ -226,10 +222,6 @@ ALIASES: Dict[str, str] = {
    "hugging-face": "huggingface",
    "huggingface-hub": "huggingface",

-    # xiaomi
-    "mimo": "xiaomi",
-    "xiaomi-mimo": "xiaomi",
-
    # Local server aliases → virtual "local" concept (resolved via user config)
    "lmstudio": "lmstudio",
    "lm-studio": "lmstudio",
@@ -250,7 +242,6 @@ _LABEL_OVERRIDES: Dict[str, str] = {
    "nous": "Nous Portal",
    "openai-codex": "OpenAI Codex",
    "copilot-acp": "GitHub Copilot ACP",
-    "xiaomi": "Xiaomi MiMo",
    "local": "Local endpoint",
 }

@@ -304,9 +304,6 @@ def _get_named_custom_provider(requested_provider: str) -> Optional[Dict[str, An
        api_mode = _parse_api_mode(entry.get("api_mode"))
        if api_mode:
            result["api_mode"] = api_mode
-        model_name = str(entry.get("model", "") or "").strip()
-        if model_name:
-            result["model"] = model_name
        return result

    return None
@@ -332,11 +329,6 @@ def _resolve_named_custom_runtime(
    # Check if a credential pool exists for this custom endpoint
    pool_result = _try_resolve_from_custom_pool(base_url, "custom", custom_provider.get("api_mode"))
    if pool_result:
-        # Propagate the model name even when using pooled credentials —
-        # the pool doesn't know about the custom_providers model field.
-        model_name = custom_provider.get("model")
-        if model_name:
-            pool_result["model"] = model_name
        return pool_result

    api_key_candidates = [
@@ -347,7 +339,7 @@ def _resolve_named_custom_runtime(
    ]
    api_key = next((candidate for candidate in api_key_candidates if has_usable_secret(candidate)), "")

-    result = {
+    return {
        "provider": "custom",
        "api_mode": custom_provider.get("api_mode")
        or _detect_api_mode_for_url(base_url)
@@ -356,11 +348,6 @@ def _resolve_named_custom_runtime(
        "api_key": api_key or "no-key-required",
        "source": f"custom_provider:{custom_provider.get('name', requested_provider)}",
    }
-    # Propagate the model name so callers can override self.model when the
-    # provider name differs from the actual model string the API expects.
-    if custom_provider.get("model"):
-        result["model"] = custom_provider["model"]
-    return result


 def _resolve_openrouter_runtime(
@@ -106,8 +106,8 @@ _DEFAULT_PROVIDER_MODELS = {
    ],
    "zai": ["glm-5", "glm-4.7", "glm-4.5", "glm-4.5-flash"],
    "kimi-coding": ["kimi-k2.5", "kimi-k2-thinking", "kimi-k2-turbo-preview"],
-    "minimax": ["MiniMax-M2.7", "MiniMax-M2.5", "MiniMax-M2.1", "MiniMax-M2"],
-    "minimax-cn": ["MiniMax-M2.7", "MiniMax-M2.5", "MiniMax-M2.1", "MiniMax-M2"],
+    "minimax": ["MiniMax-M1", "MiniMax-M1-40k", "MiniMax-M1-80k", "MiniMax-M1-128k", "MiniMax-M1-256k", "MiniMax-M2.5", "MiniMax-M2.7"],
+    "minimax-cn": ["MiniMax-M1", "MiniMax-M1-40k", "MiniMax-M1-80k", "MiniMax-M1-128k", "MiniMax-M1-256k", "MiniMax-M2.5", "MiniMax-M2.7"],
    "ai-gateway": ["anthropic/claude-opus-4.6", "anthropic/claude-sonnet-4.6", "openai/gpt-5", "google/gemini-3-flash"],
    "kilocode": ["anthropic/claude-opus-4.6", "anthropic/claude-sonnet-4.6", "openai/gpt-5.4", "google/gemini-3-pro-preview", "google/gemini-3-flash-preview"],
    "opencode-zen": ["gpt-5.4", "gpt-5.3-codex", "claude-sonnet-4-6", "gemini-3-flash", "glm-5", "kimi-k2.5", "minimax-m2.7"],
@@ -197,12 +197,24 @@ def print_header(title: str):
    print(color(f"◆ {title}", Colors.CYAN, Colors.BOLD))


-from hermes_cli.cli_output import (  # noqa: E402
-    print_error,
-    print_info,
-    print_success,
-    print_warning,
-)
+def print_info(text: str):
+    """Print info text."""
+    print(color(f"  {text}", Colors.DIM))
+
+
+def print_success(text: str):
+    """Print success message."""
+    print(color(f"✓ {text}", Colors.GREEN))
+
+
+def print_warning(text: str):
+    """Print warning message."""
+    print(color(f"⚠ {text}", Colors.YELLOW))
+
+
+def print_error(text: str):
+    """Print error message."""
+    print(color(f"✗ {text}", Colors.RED))


 def is_interactive_stdin() -> bool:
@@ -257,9 +269,80 @@ def prompt(question: str, default: str = None, password: bool = False) -> str:


 def _curses_prompt_choice(question: str, choices: list, default: int = 0) -> int:
-    """Single-select menu using curses. Delegates to curses_radiolist."""
-    from hermes_cli.curses_ui import curses_radiolist
-    return curses_radiolist(question, choices, selected=default, cancel_returns=-1)
+    """Single-select menu using curses to avoid simple_term_menu rendering bugs."""
+    try:
+        import curses
+        result_holder = [default]
+
+        def _curses_menu(stdscr):
+            curses.curs_set(0)
+            if curses.has_colors():
+                curses.start_color()
+                curses.use_default_colors()
+                curses.init_pair(1, curses.COLOR_GREEN, -1)
+                curses.init_pair(2, curses.COLOR_YELLOW, -1)
+            cursor = default
+            scroll_offset = 0
+
+            while True:
+                stdscr.clear()
+                max_y, max_x = stdscr.getmaxyx()
+
+                # Rows available for list items: rows 2..(max_y-2) inclusive.
+                visible = max(1, max_y - 3)
+
+                # Scroll the viewport so the cursor is always visible.
+                if cursor < scroll_offset:
+                    scroll_offset = cursor
+                elif cursor >= scroll_offset + visible:
+                    scroll_offset = cursor - visible + 1
+                scroll_offset = max(0, min(scroll_offset, max(0, len(choices) - visible)))
+
+                try:
+                    stdscr.addnstr(
+                        0,
+                        0,
+                        question,
+                        max_x - 1,
+                        curses.A_BOLD | (curses.color_pair(2) if curses.has_colors() else 0),
+                    )
+                except curses.error:
+                    pass
+
+                for row, i in enumerate(range(scroll_offset, min(scroll_offset + visible, len(choices)))):
+                    y = row + 2
+                    if y >= max_y - 1:
+                        break
+                    arrow = "→" if i == cursor else " "
+                    line = f" {arrow}  {choices[i]}"
+                    attr = curses.A_NORMAL
+                    if i == cursor:
+                        attr = curses.A_BOLD
+                        if curses.has_colors():
+                            attr |= curses.color_pair(1)
+                    try:
+                        stdscr.addnstr(y, 0, line, max_x - 1, attr)
+                    except curses.error:
+                        pass
+
+                stdscr.refresh()
+                key = stdscr.getch()
+                if key in (curses.KEY_UP, ord("k")):
+                    cursor = (cursor - 1) % len(choices)
+                elif key in (curses.KEY_DOWN, ord("j")):
+                    cursor = (cursor + 1) % len(choices)
+                elif key in (curses.KEY_ENTER, 10, 13):
+                    result_holder[0] = cursor
+                    return
+                elif key in (27, ord("q")):
+                    return
+
+        curses.wrapper(_curses_menu)
+        from hermes_cli.curses_ui import flush_stdin
+        flush_stdin()
+        return result_holder[0]
+    except Exception:
+        return -1



@@ -474,8 +557,6 @@ def _print_setup_summary(config: dict, hermes_home):
        tool_status.append(("Text-to-Speech (OpenAI)", True, None))
    elif tts_provider == "minimax" and get_env_value("MINIMAX_API_KEY"):
        tool_status.append(("Text-to-Speech (MiniMax)", True, None))
-    elif tts_provider == "mistral" and get_env_value("MISTRAL_API_KEY"):
-        tool_status.append(("Text-to-Speech (Mistral Voxtral)", True, None))
    elif tts_provider == "neutts":
        try:
            import importlib.util
@@ -963,7 +1044,6 @@ def _setup_tts_provider(config: dict):
        "elevenlabs": "ElevenLabs",
        "openai": "OpenAI TTS",
        "minimax": "MiniMax TTS",
-        "mistral": "Mistral Voxtral TTS",
        "neutts": "NeuTTS",
    }
    current_label = provider_labels.get(current_provider, current_provider)
@@ -984,11 +1064,10 @@ def _setup_tts_provider(config: dict):
            "ElevenLabs (premium quality, needs API key)",
            "OpenAI TTS (good quality, needs API key)",
            "MiniMax TTS (high quality with voice cloning, needs API key)",
-            "Mistral Voxtral TTS (multilingual, native Opus, needs API key)",
            "NeuTTS (local on-device, free, ~300MB model download)",
        ]
    )
-    providers.extend(["edge", "elevenlabs", "openai", "minimax", "mistral", "neutts"])
+    providers.extend(["edge", "elevenlabs", "openai", "minimax", "neutts"])
    choices.append(f"Keep current ({current_label})")
    keep_current_idx = len(choices) - 1
    idx = prompt_choice("Select TTS provider:", choices, keep_current_idx)
@@ -1066,18 +1145,6 @@ def _setup_tts_provider(config: dict):
                print_warning("No API key provided. Falling back to Edge TTS.")
                selected = "edge"

-    elif selected == "mistral":
-        existing = get_env_value("MISTRAL_API_KEY")
-        if not existing:
-            print()
-            api_key = prompt("Mistral API key for TTS", password=True)
-            if api_key:
-                save_env_value("MISTRAL_API_KEY", api_key)
-                print_success("Mistral TTS API key saved")
-            else:
-                print_warning("No API key provided. Falling back to Edge TTS.")
-                selected = "edge"
-
    # Save the selection
    if "tts" not in config:
        config["tts"] = {}
@@ -1969,48 +2036,6 @@ def _setup_weixin():
    _gateway_setup_weixin()


-def _setup_signal():
-    """Configure Signal via gateway setup."""
-    from hermes_cli.gateway import _setup_signal as _gateway_setup_signal
-    _gateway_setup_signal()
-
-
-def _setup_email():
-    """Configure Email via gateway setup."""
-    from hermes_cli.gateway import _setup_email as _gateway_setup_email
-    _gateway_setup_email()
-
-
-def _setup_sms():
-    """Configure SMS (Twilio) via gateway setup."""
-    from hermes_cli.gateway import _setup_sms as _gateway_setup_sms
-    _gateway_setup_sms()
-
-
-def _setup_dingtalk():
-    """Configure DingTalk via gateway setup."""
-    from hermes_cli.gateway import _setup_dingtalk as _gateway_setup_dingtalk
-    _gateway_setup_dingtalk()
-
-
-def _setup_feishu():
-    """Configure Feishu / Lark via gateway setup."""
-    from hermes_cli.gateway import _setup_feishu as _gateway_setup_feishu
-    _gateway_setup_feishu()
-
-
-def _setup_wecom():
-    """Configure WeCom (Enterprise WeChat) via gateway setup."""
-    from hermes_cli.gateway import _setup_wecom as _gateway_setup_wecom
-    _gateway_setup_wecom()
-
-
-def _setup_wecom_callback():
-    """Configure WeCom Callback (self-built app) via gateway setup."""
-    from hermes_cli.gateway import _setup_wecom_callback as _gw_setup
-    _gw_setup()
-
-
 def _setup_bluebubbles():
    """Configure BlueBubbles iMessage gateway."""
    print_header("BlueBubbles (iMessage)")
@@ -2127,16 +2152,9 @@ _GATEWAY_PLATFORMS = [
    ("Telegram", "TELEGRAM_BOT_TOKEN", _setup_telegram),
    ("Discord", "DISCORD_BOT_TOKEN", _setup_discord),
    ("Slack", "SLACK_BOT_TOKEN", _setup_slack),
-    ("Signal", "SIGNAL_HTTP_URL", _setup_signal),
-    ("Email", "EMAIL_ADDRESS", _setup_email),
-    ("SMS (Twilio)", "TWILIO_ACCOUNT_SID", _setup_sms),
    ("Matrix", "MATRIX_ACCESS_TOKEN", _setup_matrix),
    ("Mattermost", "MATTERMOST_TOKEN", _setup_mattermost),
    ("WhatsApp", "WHATSAPP_ENABLED", _setup_whatsapp),
-    ("DingTalk", "DINGTALK_CLIENT_ID", _setup_dingtalk),
-    ("Feishu / Lark", "FEISHU_APP_ID", _setup_feishu),
-    ("WeCom (Enterprise WeChat)", "WECOM_BOT_ID", _setup_wecom),
-    ("WeCom Callback (Self-Built App)", "WECOM_CALLBACK_CORP_ID", _setup_wecom_callback),
    ("Weixin (WeChat)", "WEIXIN_ACCOUNT_ID", _setup_weixin),
    ("BlueBubbles (iMessage)", "BLUEBUBBLES_SERVER_URL", _setup_bluebubbles),
    ("Webhooks (GitHub, GitLab, etc.)", "WEBHOOK_ENABLED", _setup_webhooks),
@@ -2178,17 +2196,10 @@ def setup_gateway(config: dict):
        get_env_value("TELEGRAM_BOT_TOKEN")
        or get_env_value("DISCORD_BOT_TOKEN")
        or get_env_value("SLACK_BOT_TOKEN")
-        or get_env_value("SIGNAL_HTTP_URL")
-        or get_env_value("EMAIL_ADDRESS")
-        or get_env_value("TWILIO_ACCOUNT_SID")
        or get_env_value("MATTERMOST_TOKEN")
        or get_env_value("MATRIX_ACCESS_TOKEN")
        or get_env_value("MATRIX_PASSWORD")
        or get_env_value("WHATSAPP_ENABLED")
-        or get_env_value("DINGTALK_CLIENT_ID")
-        or get_env_value("FEISHU_APP_ID")
-        or get_env_value("WECOM_BOT_ID")
-        or get_env_value("WEIXIN_ACCOUNT_ID")
        or get_env_value("BLUEBUBBLES_SERVER_URL")
        or get_env_value("WEBHOOK_ENABLED")
    )
@@ -2377,30 +2388,12 @@ def _get_section_config_summary(config: dict, section_key: str) -> Optional[str]
            platforms.append("Discord")
        if get_env_value("SLACK_BOT_TOKEN"):
            platforms.append("Slack")
-        if get_env_value("SIGNAL_ACCOUNT"):
-            platforms.append("Signal")
-        if get_env_value("EMAIL_ADDRESS"):
-            platforms.append("Email")
-        if get_env_value("TWILIO_ACCOUNT_SID"):
-            platforms.append("SMS")
-        if get_env_value("MATRIX_ACCESS_TOKEN") or get_env_value("MATRIX_PASSWORD"):
-            platforms.append("Matrix")
-        if get_env_value("MATTERMOST_TOKEN"):
-            platforms.append("Mattermost")
        if get_env_value("WHATSAPP_PHONE_NUMBER_ID"):
            platforms.append("WhatsApp")
-        if get_env_value("DINGTALK_CLIENT_ID"):
-            platforms.append("DingTalk")
-        if get_env_value("FEISHU_APP_ID"):
-            platforms.append("Feishu")
-        if get_env_value("WECOM_BOT_ID"):
-            platforms.append("WeCom")
-        if get_env_value("WEIXIN_ACCOUNT_ID"):
-            platforms.append("Weixin")
+        if get_env_value("SIGNAL_ACCOUNT"):
+            platforms.append("Signal")
        if get_env_value("BLUEBUBBLES_SERVER_URL"):
            platforms.append("BlueBubbles")
-        if get_env_value("WEBHOOK_ENABLED"):
-            platforms.append("Webhooks")
        if platforms:
            return ", ".join(platforms)
        return None  # No platforms configured — section must run
@@ -2929,33 +2922,19 @@ def run_setup_wizard(args):
    _offer_launch_chat()


-def _resolve_hermes_chat_argv() -> Optional[list[str]]:
-    """Resolve argv for launching ``hermes chat`` in a fresh process."""
-    hermes_bin = shutil.which("hermes")
-    if hermes_bin:
-        return [hermes_bin, "chat"]
-
-    try:
-        if importlib.util.find_spec("hermes_cli") is not None:
-            return [sys.executable, "-m", "hermes_cli.main", "chat"]
-    except Exception:
-        pass
-
-    return None
-
-
 def _offer_launch_chat():
    """Prompt the user to jump straight into chat after setup."""
    print()
-    if not prompt_yes_no("Launch hermes chat now?", True):
-        return
-
-    chat_argv = _resolve_hermes_chat_argv()
-    if not chat_argv:
-        print_info("Could not relaunch Hermes automatically. Run 'hermes chat' manually.")
-        return
-
-    os.execvp(chat_argv[0], chat_argv)
+    if prompt_yes_no("Launch hermes chat now?", True):
+        from hermes_cli.main import cmd_chat
+        from types import SimpleNamespace
+        cmd_chat(SimpleNamespace(
+            query=None, resume=None, continue_last=None, model=None,
+            provider=None, effort=None, skin=None, oneshot=False,
+            quiet=False, verbose=False, toolsets=None, skills=None,
+            yolo=False, source=None, worktree=False, checkpoints=False,
+            pass_session_id=False, max_turns=None,
+        ))


 def _run_first_time_quick_setup(config: dict, hermes_home, is_existing: bool):
@@ -15,12 +15,25 @@ from typing import List, Optional, Set

 from hermes_cli.config import load_config, save_config
 from hermes_cli.colors import Colors, color
-from hermes_cli.platforms import PLATFORMS as _PLATFORMS, platform_label

-# Backward-compatible view: {key: label_string} so existing code that
-# iterates ``PLATFORMS.items()`` or calls ``PLATFORMS.get(key)`` keeps
-# working without changes to every call site.
-PLATFORMS = {k: info.label for k, info in _PLATFORMS.items() if k != "api_server"}
+PLATFORMS = {
+    "cli":      "🖥️  CLI",
+    "telegram": "📱 Telegram",
+    "discord":  "💬 Discord",
+    "slack":    "💼 Slack",
+    "whatsapp": "📱 WhatsApp",
+    "signal":   "📡 Signal",
+    "bluebubbles": "💬 BlueBubbles",
+    "email":    "📧 Email",
+    "homeassistant": "🏠 Home Assistant",
+    "mattermost": "💬 Mattermost",
+    "matrix":   "💬 Matrix",
+    "dingtalk": "💬 DingTalk",
+    "feishu": "🪽 Feishu",
+    "wecom": "💬 WeCom",
+    "weixin": "💬 Weixin",
+    "webhook": "🔗 Webhook",
+}

 # ─── Config Helpers ───────────────────────────────────────────────────────────

@@ -141,8 +141,11 @@ def show_status(args):
        display = redact_key(value) if not show_all else value
        print(f"  {name:<12}  {check_mark(has_key)} {display}")

-    from hermes_cli.auth import get_anthropic_key
-    anthropic_value = get_anthropic_key()
+    anthropic_value = (
+        get_env_value("ANTHROPIC_TOKEN")
+        or get_env_value("ANTHROPIC_API_KEY")
+        or ""
+    )
    anthropic_display = redact_key(anthropic_value) if not show_all else anthropic_value
    print(f"  {'Anthropic':<12}  {check_mark(bool(anthropic_value))} {anthropic_display}")

@@ -302,7 +305,6 @@ def show_status(args):
        "DingTalk": ("DINGTALK_CLIENT_ID", None),
        "Feishu": ("FEISHU_APP_ID", "FEISHU_HOME_CHANNEL"),
        "WeCom": ("WECOM_BOT_ID", "WECOM_HOME_CHANNEL"),
-        "WeCom Callback": ("WECOM_CALLBACK_CORP_ID", None),
        "Weixin": ("WEIXIN_ACCOUNT_ID", "WEIXIN_HOME_CHANNEL"),
        "BlueBubbles": ("BLUEBUBBLES_SERVER_URL", "BLUEBUBBLES_HOME_CHANNEL"),
    }
@@ -1,351 +0,0 @@
-"""Random tips shown at CLI session start to help users discover features."""
-
-import random
-from typing import Optional
-
-# ---------------------------------------------------------------------------
-# Tip corpus — one-liners covering slash commands, CLI flags, config,
-# keybindings, tools, gateway, skills, profiles, and workflow tricks.
-# ---------------------------------------------------------------------------
-
-TIPS = [
-    # --- Slash Commands ---
-    "/btw <question> asks a quick side question without tools or history — great for clarifications.",
-    "/background <prompt> runs a task in a separate session while your current one stays free.",
-    "/branch forks the current session so you can explore a different direction without losing progress.",
-    "/compress manually compresses conversation context when things get long.",
-    "/rollback lists filesystem checkpoints — restore files the agent modified to any prior state.",
-    "/rollback diff 2 previews what changed since checkpoint 2 without restoring anything.",
-    "/rollback 2 src/file.py restores a single file from a specific checkpoint.",
-    "/title \"my project\" names your session — resume it later with /resume or hermes -c.",
-    "/resume picks up where you left off in a previously named session.",
-    "/queue <prompt> queues a message for the next turn without interrupting the current one.",
-    "/undo removes the last user/assistant exchange from the conversation.",
-    "/retry resends your last message — useful when the agent's response wasn't quite right.",
-    "/verbose cycles tool progress display: off → new → all → verbose.",
-    "/reasoning high increases the model's thinking depth. /reasoning show displays the reasoning.",
-    "/fast toggles priority processing for faster API responses (provider-dependent).",
-    "/yolo skips all dangerous command approval prompts for the rest of the session.",
-    "/model lets you switch models mid-session — try /model sonnet or /model gpt-5.",
-    "/model --global changes your default model permanently.",
-    "/personality pirate sets a fun personality — 14 built-in options from kawaii to shakespeare.",
-    "/skin changes the CLI theme — try ares, mono, slate, poseidon, or charizard.",
-    "/statusbar toggles a persistent bar showing model, tokens, context fill %, cost, and duration.",
-    "/tools disable browser temporarily removes browser tools for the current session.",
-    "/browser connect attaches browser tools to your running Chrome instance via CDP.",
-    "/plugins lists installed plugins and their status.",
-    "/cron manages scheduled tasks — set up recurring prompts with delivery to any platform.",
-    "/reload-mcp hot-reloads MCP server configuration without restarting.",
-    "/usage shows token usage, cost breakdown, and session duration.",
-    "/insights shows usage analytics for the last 30 days.",
-    "/paste checks your clipboard for an image and attaches it to your next message.",
-    "/profile shows which profile is active and its home directory.",
-    "/config shows your current configuration at a glance.",
-    "/stop kills all running background processes spawned by the agent.",
-
-    # --- @ Context References ---
-    "@file:path/to/file.py injects file contents directly into your message.",
-    "@file:main.py:10-50 injects only lines 10-50 of a file.",
-    "@folder:src/ injects a directory tree listing.",
-    "@diff injects your unstaged git changes into the message.",
-    "@staged injects your staged git changes (git diff --staged).",
-    "@git:5 injects the last 5 commits with full patches.",
-    "@url:https://example.com fetches and injects a web page's content.",
-    "Typing @ triggers filesystem path completion — navigate to any file interactively.",
-    "Combine multiple references: \"Review @file:main.py and @file:test.py for consistency.\"",
-
-    # --- Keybindings ---
-    "Alt+Enter (or Ctrl+J) inserts a newline for multi-line input.",
-    "Ctrl+C interrupts the agent. Double-press within 2 seconds to force exit.",
-    "Ctrl+Z suspends Hermes to the background — run fg in your shell to resume.",
-    "Tab accepts auto-suggestion ghost text or autocompletes slash commands.",
-    "Type a new message while the agent is working to interrupt and redirect it.",
-    "Alt+V pastes an image from your clipboard into the conversation.",
-    "Pasting 5+ lines auto-saves to a file and inserts a compact reference instead.",
-
-    # --- CLI Flags ---
-    "hermes -c resumes your most recent CLI session. hermes -c \"project name\" resumes by title.",
-    "hermes -w creates an isolated git worktree — perfect for parallel agent workflows.",
-    "hermes -w -q \"Fix issue #42\" combines worktree isolation with a one-shot query.",
-    "hermes chat -t web,terminal enables only specific toolsets for a focused session.",
-    "hermes chat -s github-pr-workflow preloads a skill at launch.",
-    "hermes chat -q \"query\" runs a single non-interactive query and exits.",
-    "hermes chat --max-turns 200 overrides the default 90-iteration limit per turn.",
-    "hermes chat --checkpoints enables filesystem snapshots before every destructive file change.",
-    "hermes --yolo bypasses all dangerous command approval prompts for the entire session.",
-    "hermes chat --source telegram tags the session for filtering in hermes sessions list.",
-    "hermes -p work chat runs under a specific profile without changing your default.",
-
-    # --- CLI Subcommands ---
-    "hermes doctor --fix diagnoses and auto-repairs config and dependency issues.",
-    "hermes dump outputs a compact setup summary — great for bug reports.",
-    "hermes config set KEY VALUE auto-routes secrets to .env and everything else to config.yaml.",
-    "hermes config edit opens config.yaml in your default editor.",
-    "hermes config check scans for missing or stale configuration options.",
-    "hermes sessions browse opens an interactive session picker with search.",
-    "hermes sessions stats shows session counts by platform and database size.",
-    "hermes sessions prune --older-than 30 cleans up old sessions.",
-    "hermes skills search react --source skills-sh searches the skills.sh public directory.",
-    "hermes skills check scans installed hub skills for upstream updates.",
-    "hermes skills tap add myorg/skills-repo adds a custom GitHub skill source.",
-    "hermes skills snapshot export setup.json exports your skill configuration for backup or sharing.",
-    "hermes mcp add github --command npx adds MCP servers from the command line.",
-    "hermes mcp serve runs Hermes itself as an MCP server for other agents.",
-    "hermes auth add lets you add multiple API keys for credential pool rotation.",
-    "hermes completion bash >> ~/.bashrc enables tab completion for all commands and profiles.",
-    "hermes logs -f follows agent.log in real time. --level WARNING --since 1h filters output.",
-    "hermes backup creates a zip backup of your entire Hermes home directory.",
-    "hermes profile create coder creates an isolated profile that becomes its own command.",
-    "hermes profile create work --clone copies your current config and keys to a new profile.",
-    "hermes update syncs new bundled skills to ALL profiles automatically.",
-    "hermes gateway install sets up Hermes as a system service (systemd/launchd).",
-    "hermes memory setup lets you configure an external memory provider (Honcho, Mem0, etc.).",
-    "hermes webhook subscribe creates event-driven webhook routes with HMAC validation.",
-
-    # --- Configuration ---
-    "Set display.bell_on_complete: true in config.yaml to hear a bell when long tasks finish.",
-    "Set display.streaming: true to see tokens appear in real time as the model generates.",
-    "Set display.show_reasoning: true to watch the model's chain-of-thought reasoning.",
-    "Set display.compact: true to reduce whitespace in output for denser information.",
-    "Set display.busy_input_mode: queue to queue messages instead of interrupting the agent.",
-    "Set display.resume_display: minimal to skip the full conversation recap on session resume.",
-    "Set compression.threshold: 0.50 to control when auto-compression fires (default: 50% of context).",
-    "Set agent.max_turns: 200 to let the agent take more tool-calling steps per turn.",
-    "Set file_read_max_chars: 200000 to increase the max content per read_file call.",
-    "Set approvals.mode: smart to let an LLM auto-approve safe commands and auto-deny dangerous ones.",
-    "Set fallback_model in config.yaml to automatically fail over to a backup provider.",
-    "Set privacy.redact_pii: true to hash user IDs and phone numbers before sending to the LLM.",
-    "Set browser.record_sessions: true to auto-record browser sessions as WebM videos.",
-    "Set worktree: true in config.yaml to always create a git worktree (same as hermes -w).",
-    "Set security.website_blocklist.enabled: true to block specific domains from web tools.",
-    "Set cron.wrap_response: false to deliver raw agent output without the cron header/footer.",
-    "HERMES_TIMEZONE overrides the server timezone with any IANA timezone string.",
-    "Environment variable substitution works in config.yaml: use ${VAR_NAME} syntax.",
-    "Quick commands in config.yaml run shell commands instantly with zero token usage.",
-    "Custom personalities can be defined in config.yaml under agent.personalities.",
-    "provider_routing controls OpenRouter provider sorting, whitelisting, and blacklisting.",
-
-    # --- Tools & Capabilities ---
-    "execute_code runs Python scripts that call Hermes tools programmatically — results stay out of context.",
-    "delegate_task spawns up to 3 concurrent sub-agents with isolated contexts for parallel work.",
-    "web_extract works on PDF URLs — pass any PDF link and it converts to markdown.",
-    "search_files is ripgrep-backed and faster than grep — use it instead of terminal grep.",
-    "patch uses 9 fuzzy matching strategies so minor whitespace differences won't break edits.",
-    "patch supports V4A format for bulk multi-file edits in a single call.",
-    "read_file suggests similar filenames when a file isn't found.",
-    "read_file auto-deduplicates — re-reading an unchanged file returns a lightweight stub.",
-    "browser_vision takes a screenshot and analyzes it with AI — works for CAPTCHAs and visual content.",
-    "browser_console can evaluate JavaScript expressions in the page context.",
-    "image_generate creates images with FLUX 2 Pro and automatic 2x upscaling.",
-    "text_to_speech converts text to audio — plays as voice bubbles on Telegram.",
-    "send_message can reach any connected messaging platform from within a session.",
-    "The todo tool helps the agent track complex multi-step tasks during a session.",
-    "session_search performs full-text search across ALL past conversations.",
-    "The agent automatically saves preferences, corrections, and environment facts to memory.",
-    "mixture_of_agents routes hard problems through 4 frontier LLMs collaboratively.",
-    "Terminal commands support background mode with notify_on_complete for long-running tasks.",
-    "Terminal background processes support watch_patterns to alert on specific output lines.",
-    "The terminal tool supports 6 backends: local, Docker, SSH, Modal, Daytona, and Singularity.",
-
-    # --- Profiles ---
-    "Each profile gets its own config, API keys, memory, sessions, skills, and cron jobs.",
-    "Profile names become shell commands — 'hermes profile create coder' creates the 'coder' command.",
-    "hermes profile export coder -o backup.tar.gz creates a portable profile archive.",
-    "If two profiles accidentally share a bot token, the second gateway is blocked with a clear error.",
-
-    # --- Sessions ---
-    "Sessions auto-generate descriptive titles after the first exchange — no manual naming needed.",
-    "Session titles support lineage: \"my project\" → \"my project #2\" → \"my project #3\".",
-    "When exiting, Hermes prints a resume command with session ID and stats.",
-    "hermes sessions export backup.jsonl exports all sessions for backup or analysis.",
-    "hermes -r SESSION_ID resumes any specific past session by its ID.",
-
-    # --- Memory ---
-    "Memory is a frozen snapshot — changes appear in the system prompt only at next session start.",
-    "Memory entries are automatically scanned for prompt injection and exfiltration patterns.",
-    "The agent has two memory stores: personal notes (~2200 chars) and user profile (~1375 chars).",
-    "Corrections you give the agent (\"no, do it this way\") are often auto-saved to memory.",
-
-    # --- Skills ---
-    "Over 80 bundled skills covering github, creative, mlops, productivity, research, and more.",
-    "Every installed skill automatically becomes a slash command — type / to see them all.",
-    "hermes skills install official/security/1password installs optional skills from the repo.",
-    "Skills can restrict to specific OS platforms — some only load on macOS or Linux.",
-    "skills.external_dirs in config.yaml lets you load skills from custom directories.",
-    "The agent can create its own skills as procedural memory using skill_manage.",
-    "The plan skill saves markdown plans under .hermes/plans/ in the active workspace.",
-
-    # --- Cron & Scheduling ---
-    "Cron jobs can attach skills: hermes cron add --skill blogwatcher \"Check for new posts\".",
-    "Cron delivery targets include telegram, discord, slack, email, sms, and 12+ more platforms.",
-    "If a cron response starts with [SILENT], delivery is suppressed — useful for monitoring-only jobs.",
-    "Cron supports relative delays (30m), intervals (every 2h), cron expressions, and ISO timestamps.",
-    "Cron jobs run in completely fresh agent sessions — prompts must be self-contained.",
-
-    # --- Voice ---
-    "Voice mode works with zero API keys if faster-whisper is installed (free local speech-to-text).",
-    "Five TTS providers available: Edge TTS (free), ElevenLabs, OpenAI, NeuTTS (free local), MiniMax.",
-    "/voice on enables voice mode in the CLI. Ctrl+B toggles push-to-talk recording.",
-    "Streaming TTS plays sentences as they generate — you don't wait for the full response.",
-    "Voice messages on Telegram, Discord, WhatsApp, and Slack are auto-transcribed.",
-
-    # --- Gateway & Messaging ---
-    "Hermes runs on 18 platforms: Telegram, Discord, Slack, WhatsApp, Signal, Matrix, email, and more.",
-    "hermes gateway install sets it up as a system service that starts on boot.",
-    "DingTalk uses Stream Mode — no webhooks or public URL needed.",
-    "BlueBubbles brings iMessage to Hermes via a local macOS server.",
-    "Webhook routes support HMAC validation, rate limiting, and event filtering.",
-    "The API server exposes an OpenAI-compatible endpoint compatible with Open WebUI and LibreChat.",
-    "Discord voice channel mode: the bot joins VC, transcribes speech, and talks back.",
-    "group_sessions_per_user: true gives each person their own session in group chats.",
-    "/sethome marks a chat as the home channel for cron job deliveries.",
-    "The gateway supports inactivity-based timeouts — active agents can run indefinitely.",
-
-    # --- Security ---
-    "Dangerous command approval has 4 tiers: once, session, always (permanent allowlist), deny.",
-    "Smart approval mode uses an LLM to auto-approve safe commands and flag dangerous ones.",
-    "SSRF protection blocks private networks, loopback, link-local, and cloud metadata addresses.",
-    "Tirith pre-exec scanning detects homograph URL spoofing and pipe-to-interpreter patterns.",
-    "MCP subprocesses receive a filtered environment — only safe system vars pass through.",
-    "Context files (.hermes.md, AGENTS.md) are security-scanned for prompt injection before loading.",
-    "command_allowlist in config.yaml permanently approves specific shell command patterns.",
-
-    # --- Context & Compression ---
-    "Context auto-compresses when it reaches the threshold — memories are flushed and history summarized.",
-    "The status bar turns yellow, then orange, then red as context fills up.",
-    "SOUL.md at ~/.hermes/SOUL.md is the agent's primary identity — customize it to shape behavior.",
-    "Hermes loads project context from .hermes.md, AGENTS.md, CLAUDE.md, or .cursorrules (first match).",
-    "Subdirectory AGENTS.md files are discovered progressively as the agent navigates into folders.",
-    "Context files are capped at 20,000 characters with smart head/tail truncation.",
-
-    # --- Browser ---
-    "Five browser providers: local Chromium, Browserbase, Browser Use, Camofox, and Firecrawl.",
-    "Camofox is an anti-detection browser — Firefox fork with C++ fingerprint spoofing.",
-    "browser_navigate returns a page snapshot automatically — no need to call browser_snapshot after.",
-    "browser_vision with annotate=true overlays numbered labels on interactive elements.",
-
-    # --- MCP ---
-    "MCP servers are configured in config.yaml — both stdio and HTTP transports supported.",
-    "Per-server tool filtering: tools.include whitelists and tools.exclude blacklists specific tools.",
-    "MCP servers auto-generate toolsets at runtime — hermes tools can toggle them per platform.",
-    "MCP OAuth support: auth: oauth enables browser-based authorization with PKCE.",
-
-    # --- Checkpoints & Rollback ---
-    "Checkpoints have zero overhead when no files are modified — enabled by default.",
-    "A pre-rollback snapshot is saved automatically so you can undo the undo.",
-    "/rollback also undoes the conversation turn, so the agent doesn't remember rolled-back changes.",
-    "Checkpoints use shadow repos in ~/.hermes/checkpoints/ — your project's .git is never touched.",
-
-    # --- Batch & Data ---
-    "batch_runner.py processes hundreds of prompts in parallel for training data generation.",
-    "hermes chat -Q enables quiet mode for programmatic use — suppresses banner and spinner.",
-    "Trajectory saving (--save-trajectories) captures full tool-use traces for model training.",
-
-    # --- Plugins ---
-    "Three plugin types: general (tools/hooks), memory providers, and context engines.",
-    "hermes plugins install owner/repo installs plugins directly from GitHub.",
-    "8 external memory providers available: Honcho, OpenViking, Mem0, Hindsight, and more.",
-    "Plugin hooks include pre_tool_call, post_tool_call, pre_llm_call, and post_llm_call.",
-
-    # --- Miscellaneous ---
-    "Prompt caching (Anthropic) reduces costs by reusing cached system prompt prefixes.",
-    "The agent auto-generates session titles in a background thread — zero latency impact.",
-    "Smart model routing can auto-route simple queries to a cheaper model.",
-    "Slash commands support prefix matching: /h resolves to /help, /mod to /model.",
-    "Dragging a file path into the terminal auto-attaches images or sends as context.",
-    ".worktreeinclude in your repo root lists gitignored files to copy into worktrees.",
-    "hermes acp runs Hermes as an ACP server for VS Code, Zed, and JetBrains integration.",
-    "Custom providers: save named endpoints in config.yaml under custom_providers.",
-    "HERMES_EPHEMERAL_SYSTEM_PROMPT injects a system prompt that's never persisted to history.",
-    "credential_pool_strategies supports fill_first, round_robin, least_used, and random rotation.",
-    "hermes login supports OAuth-based auth for Nous and OpenAI Codex providers.",
-    "The API server supports both Chat Completions and Responses API with server-side state.",
-    "tool_preview_length: 0 in config shows full file paths in the spinner's activity feed.",
-    "hermes status --deep runs deeper diagnostic checks across all components.",
-
-    # --- Hidden Gems & Power-User Tricks ---
-    "BOOT.md at ~/.hermes/BOOT.md runs automatically on every gateway start — use it for startup checks.",
-    "Cron jobs can attach a Python script (--script) whose stdout is injected into the prompt as context.",
-    "Cron scripts live in ~/.hermes/scripts/ and run before the agent — perfect for data collection pipelines.",
-    "prefill_messages_file in config.yaml injects few-shot examples into every API call, never saved to history.",
-    "SOUL.md completely replaces the agent's default identity — rewrite it to make Hermes your own.",
-    "SOUL.md is auto-seeded with a default personality on first run. Edit ~/.hermes/SOUL.md to customize.",
-    "/compress <focus topic> allocates 60-70% of the summary budget to your topic and aggressively trims the rest.",
-    "On second+ compression, the compressor updates the previous summary instead of starting from scratch.",
-    "Before a gateway session reset, Hermes auto-flushes important facts to memory in the background.",
-    "network.force_ipv4: true in config.yaml fixes hangs on servers with broken IPv6 — monkey-patches socket.",
-    "The terminal tool annotates common exit codes: grep returning 1 = 'No matches found (not an error)'.",
-    "Failed foreground terminal commands auto-retry up to 3 times with exponential backoff (2s, 4s, 8s).",
-    "Bare sudo commands are auto-rewritten to pipe SUDO_PASSWORD from .env — no interactive prompt needed.",
-    "execute_code has built-in helpers: json_parse() for tolerant parsing, shell_quote(), and retry() with backoff.",
-    "execute_code's 7 sandbox tools (web_search, terminal, read/write/search/patch) use RPC — never enter context.",
-    "Reading the same file region 3+ times triggers a warning. At 4+, it's hard-blocked to prevent loops.",
-    "write_file and patch detect if a file was externally modified since the last read and warn about staleness.",
-    "V4A patch format supports Add File, Delete File, and Move File directives — not just Update.",
-    "MCP servers can request LLM completions back via sampling — the agent becomes a tool for the server.",
-    "MCP servers send notifications/tools/list_changed to trigger automatic tool re-registration without restart.",
-    "delegate_task with acp_command: 'claude' spawns Claude Code as a child agent from any platform.",
-    "Delegation has a heartbeat thread — child activity propagates to the parent, preventing gateway timeouts.",
-    "When a provider returns HTTP 402 (payment required), the auxiliary client auto-falls back to the next one.",
-    "agent.tool_use_enforcement steers models that describe actions instead of calling tools — auto for GPT/Codex.",
-    "agent.restart_drain_timeout (default 60s) lets running agents finish before a gateway restart takes effect.",
-    "The gateway caches AIAgent instances per session — destroying this cache breaks Anthropic prompt caching.",
-    "Any website can expose skills via /.well-known/skills/index.json — the skills hub discovers them automatically.",
-    "The skills audit log at ~/.hermes/skills/.hub/audit.log tracks every install and removal operation.",
-    "Stale git worktrees are auto-cleaned: 24-72h old with no unpushed commits get pruned on startup.",
-    "Each profile gets its own subprocess HOME at HERMES_HOME/home/ — isolated git, ssh, npm, gh configs.",
-    "HERMES_HOME_MODE env var (octal, e.g. 0701) sets custom directory permissions for web server traversal.",
-    "Container mode: place .container-mode in HERMES_HOME and the host CLI auto-execs into the container.",
-    "Ctrl+C has 5 priority tiers: cancel recording → cancel prompts → cancel picker → interrupt agent → exit.",
-    "Every interrupt during an agent run is logged to ~/.hermes/interrupt_debug.log with timestamps.",
-    "BROWSER_CDP_URL connects browser tools to any running Chrome — accepts WebSocket, HTTP, or host:port.",
-    "BROWSERBASE_ADVANCED_STEALTH=true enables advanced anti-detection with custom Chromium (Scale Plan).",
-    "The CLI auto-switches to compact mode in terminals narrower than 80 columns.",
-    "Quick commands support two types: exec (run shell command directly) and alias (redirect to another command).",
-    "Per-task delegation model: delegation.model and delegation.provider in config route subagents to cheaper models.",
-    "delegation.reasoning_effort independently controls thinking depth for subagents.",
-    "display.platforms in config.yaml allows per-platform display overrides: {telegram: {tool_progress: all}}.",
-    "human_delay.mode in config simulates human typing speed — configurable min_ms/max_ms range.",
-    "Config version migrations run automatically on load — new config keys appear without manual intervention.",
-    "GPT and Codex models get special system prompt guidance for tool discipline and mandatory tool use.",
-    "Gemini models get tailored directives for absolute paths, parallel tool calls, and non-interactive commands.",
-    "context.engine in config.yaml can be set to a plugin name for alternative context management strategies.",
-    "Browser pages over 8000 tokens are auto-summarized by the auxiliary LLM before returning to the agent.",
-    "The compressor does a cheap pre-pass: tool outputs over 200 chars are replaced with placeholders before the LLM runs.",
-    "When compression fails, further attempts are paused for 10 minutes to avoid API hammering.",
-    "Long dangerous commands (>70 chars) get a 'view' option in the approval prompt to see the full text first.",
-    "Audio level visualization shows ▁▂▃▄▅▆▇ bars during voice recording based on microphone RMS levels.",
-    "Profile names cannot collide with existing PATH binaries — 'hermes profile create ls' would be rejected.",
-    "hermes profile create backup --clone-all copies everything (config, keys, SOUL.md, memories, skills, sessions).",
-    "The voice record key is configurable via voice.record_key in config.yaml — not just Ctrl+B.",
-    ".cursorrules and .cursor/rules/*.mdc files are auto-detected and loaded as project context.",
-    "Context files support 10+ prompt injection patterns — invisible Unicode, 'ignore instructions', exfil attempts.",
-    "GPT-5 and Codex use 'developer' role instead of 'system' in the message format.",
-    "Per-task auxiliary overrides: auxiliary.vision.provider, auxiliary.compression.model, etc. in config.yaml.",
-    "The auxiliary client treats 'main' as a provider alias — resolves to your actual primary provider + model.",
-    "Smart routing can auto-route simple queries to a cheaper model — set smart_model_routing.enabled: true.",
-    "hermes claw migrate --dry-run previews OpenClaw migration without writing anything.",
-    "File paths pasted with quotes or escaped spaces are handled automatically — no manual cleanup needed.",
-    "Slash commands never trigger the large-paste collapse — /command with big arguments works correctly.",
-    "In interrupt mode, slash commands typed during agent execution bypass interrupt logic and run immediately.",
-    "HERMES_DEV=1 bypasses container mode detection for local development.",
-    "Each MCP server gets its own toolset (mcp-servername) that can be toggled independently via hermes tools.",
-    "MCP ${ENV_VAR} placeholders in config are resolved at server spawn — including vars from ~/.hermes/.env.",
-    "Skills from trusted repos (NousResearch) get a 'trusted' security level; community skills get extra scanning.",
-    "The skills quarantine at ~/.hermes/skills/.hub/quarantine/ holds skills pending security review.",
-]
-
-
-def get_random_tip(exclude_recent: int = 0) -> str:
-    """Return a random tip string.
-
-    Args:
-        exclude_recent: not used currently; reserved for future
-            deduplication across sessions.
-    """
-    return random.choice(TIPS)
-
-
-def get_tip_count() -> int:
-    """Return the total number of tips available."""
-    return len(TIPS)
@@ -33,13 +33,33 @@ PROJECT_ROOT = Path(__file__).parent.parent.resolve()

 # ─── UI Helpers (shared with setup.py) ────────────────────────────────────────

-from hermes_cli.cli_output import (  # noqa: E402 — late import block
-    print_error as _print_error,
-    print_info as _print_info,
-    print_success as _print_success,
-    print_warning as _print_warning,
-    prompt as _prompt,
-)
+def _print_info(text: str):
+    print(color(f"  {text}", Colors.DIM))
+
+def _print_success(text: str):
+    print(color(f"✓ {text}", Colors.GREEN))
+
+def _print_warning(text: str):
+    print(color(f"⚠ {text}", Colors.YELLOW))
+
+def _print_error(text: str):
+    print(color(f"✗ {text}", Colors.RED))
+
+def _prompt(question: str, default: str = None, password: bool = False) -> str:
+    if default:
+        display = f"{question} [{default}]: "
+    else:
+        display = f"{question}: "
+    try:
+        if password:
+            import getpass
+            value = getpass.getpass(color(display, Colors.YELLOW))
+        else:
+            value = input(color(display, Colors.YELLOW))
+        return value.strip() or default or ""
+    except (KeyboardInterrupt, EOFError):
+        print()
+        return default or ""

 # ─── Toolset Registry ─────────────────────────────────────────────────────────

@@ -98,14 +118,25 @@ def _get_plugin_toolset_keys() -> set:
    except Exception:
        return set()

-# Platform display config — derived from the canonical registry so every
-# module shares the same data.  Kept as dict-of-dicts for backward
-# compatibility with existing ``PLATFORMS[key]["label"]`` access patterns.
-from hermes_cli.platforms import PLATFORMS as _PLATFORMS_REGISTRY
-
+# Platform display config
 PLATFORMS = {
-    k: {"label": info.label, "default_toolset": info.default_toolset}
-    for k, info in _PLATFORMS_REGISTRY.items()
+    "cli":      {"label": "🖥️  CLI",       "default_toolset": "hermes-cli"},
+    "telegram": {"label": "📱 Telegram",   "default_toolset": "hermes-telegram"},
+    "discord":  {"label": "💬 Discord",    "default_toolset": "hermes-discord"},
+    "slack":    {"label": "💼 Slack",      "default_toolset": "hermes-slack"},
+    "whatsapp": {"label": "📱 WhatsApp",   "default_toolset": "hermes-whatsapp"},
+    "signal":   {"label": "📡 Signal",     "default_toolset": "hermes-signal"},
+    "bluebubbles": {"label": "💙 BlueBubbles", "default_toolset": "hermes-bluebubbles"},
+    "homeassistant": {"label": "🏠 Home Assistant", "default_toolset": "hermes-homeassistant"},
+    "email":    {"label": "📧 Email",      "default_toolset": "hermes-email"},
+    "matrix":   {"label": "💬 Matrix",     "default_toolset": "hermes-matrix"},
+ "dingtalk": {"label": "💬 DingTalk", "default_toolset": "hermes-dingtalk"},
+    "feishu": {"label": "🪽 Feishu", "default_toolset": "hermes-feishu"},
+    "wecom": {"label": "💬 WeCom", "default_toolset": "hermes-wecom"},
+    "weixin": {"label": "💬 Weixin", "default_toolset": "hermes-weixin"},
+    "api_server": {"label": "🌐 API Server", "default_toolset": "hermes-api-server"},
+    "mattermost": {"label": "💬 Mattermost", "default_toolset": "hermes-mattermost"},
+    "webhook": {"label": "🔗 Webhook", "default_toolset": "hermes-webhook"},
 }


@@ -150,14 +181,6 @@ TOOL_CATEGORIES = {
                ],
                "tts_provider": "elevenlabs",
            },
-            {
-                "name": "Mistral (Voxtral TTS)",
-                "tag": "Multilingual, native Opus, needs MISTRAL_API_KEY",
-                "env_vars": [
-                    {"key": "MISTRAL_API_KEY", "prompt": "Mistral API key", "url": "https://console.mistral.ai/"},
-                ],
-                "tts_provider": "mistral",
-            },
        ],
    },
    "web": {
@@ -478,10 +501,6 @@ def _get_platform_tools(
        default_ts = PLATFORMS[platform]["default_toolset"]
        toolset_names = [default_ts]

-    # YAML may parse bare numeric names (e.g. ``12306:``) as int.
-    # Normalise to str so downstream sorted() never mixes types.
-    toolset_names = [str(ts) for ts in toolset_names]
-
    configurable_keys = {ts_key for ts_key, _, _ in CONFIGURABLE_TOOLSETS}

    # If the saved list contains any configurable keys directly, the user
@@ -540,7 +559,7 @@ def _get_platform_tools(
    # Special sentinel: "no_mcp" in the toolset list disables all MCP servers.
    mcp_servers = config.get("mcp_servers") or {}
    enabled_mcp_servers = {
-        str(name)
+        name
        for name, server_cfg in mcp_servers.items()
        if isinstance(server_cfg, dict)
        and _parse_enabled_flag(server_cfg.get("enabled", True), default=True)
@@ -646,9 +665,86 @@ def _toolset_has_keys(ts_key: str, config: dict = None) -> bool:
 # ─── Menu Helpers ─────────────────────────────────────────────────────────────

 def _prompt_choice(question: str, choices: list, default: int = 0) -> int:
-    """Single-select menu (arrow keys). Delegates to curses_radiolist."""
-    from hermes_cli.curses_ui import curses_radiolist
-    return curses_radiolist(question, choices, selected=default, cancel_returns=default)
+    """Single-select menu (arrow keys). Uses curses to avoid simple_term_menu
+    rendering bugs in tmux, iTerm, and other non-standard terminals."""
+
+    # Curses-based single-select — works in tmux, iTerm, and standard terminals
+    try:
+        import curses
+        result_holder = [default]
+
+        def _curses_menu(stdscr):
+            curses.curs_set(0)
+            if curses.has_colors():
+                curses.start_color()
+                curses.use_default_colors()
+                curses.init_pair(1, curses.COLOR_GREEN, -1)
+                curses.init_pair(2, curses.COLOR_YELLOW, -1)
+            cursor = default
+
+            while True:
+                stdscr.clear()
+                max_y, max_x = stdscr.getmaxyx()
+                try:
+                    stdscr.addnstr(0, 0, question, max_x - 1,
+                                   curses.A_BOLD | (curses.color_pair(2) if curses.has_colors() else 0))
+                except curses.error:
+                    pass
+
+                for i, c in enumerate(choices):
+                    y = i + 2
+                    if y >= max_y - 1:
+                        break
+                    arrow = "→" if i == cursor else " "
+                    line = f" {arrow}  {c}"
+                    attr = curses.A_NORMAL
+                    if i == cursor:
+                        attr = curses.A_BOLD
+                        if curses.has_colors():
+                            attr |= curses.color_pair(1)
+                    try:
+                        stdscr.addnstr(y, 0, line, max_x - 1, attr)
+                    except curses.error:
+                        pass
+
+                stdscr.refresh()
+                key = stdscr.getch()
+
+                if key in (curses.KEY_UP, ord('k')):
+                    cursor = (cursor - 1) % len(choices)
+                elif key in (curses.KEY_DOWN, ord('j')):
+                    cursor = (cursor + 1) % len(choices)
+                elif key in (curses.KEY_ENTER, 10, 13):
+                    result_holder[0] = cursor
+                    return
+                elif key in (27, ord('q')):
+                    return
+
+        curses.wrapper(_curses_menu)
+        from hermes_cli.curses_ui import flush_stdin
+        flush_stdin()
+        return result_holder[0]
+
+    except Exception:
+        pass
+
+    # Fallback: numbered input (Windows without curses, etc.)
+    print(color(question, Colors.YELLOW))
+    for i, c in enumerate(choices):
+        marker = "●" if i == default else "○"
+        style = Colors.GREEN if i == default else ""
+        print(color(f"  {marker} {i+1}. {c}", style) if style else f"  {marker} {i+1}. {c}")
+    while True:
+        try:
+            val = input(color(f"  Select [1-{len(choices)}] ({default + 1}): ", Colors.DIM))
+            if not val:
+                return default
+            idx = int(val) - 1
+            if 0 <= idx < len(choices):
+                return idx
+        except (ValueError, KeyboardInterrupt, EOFError):
+            print()
+            return default


 # ─── Token Estimation ────────────────────────────────────────────────────────
@@ -0,0 +1,398 @@
+"""
+Hermes Agent — Web UI server.
+
+Provides a FastAPI backend serving the Vite/React frontend and REST API
+endpoints for managing configuration, environment variables, and sessions.
+
+Usage:
+    python -m hermes_cli.main web          # Start on http://127.0.0.1:9119
+    python -m hermes_cli.main web --port 8080
+"""
+
+import os
+import sys
+import time
+from pathlib import Path
+from typing import Any, Dict, List, Optional
+
+PROJECT_ROOT = Path(__file__).parent.parent.resolve()
+if str(PROJECT_ROOT) not in sys.path:
+    sys.path.insert(0, str(PROJECT_ROOT))
+
+from hermes_cli import __version__, __release_date__
+from hermes_cli.config import (
+    DEFAULT_CONFIG,
+    OPTIONAL_ENV_VARS,
+    get_config_path,
+    get_env_path,
+    get_hermes_home,
+    load_config,
+    load_env,
+    save_config,
+    save_env_value,
+    delete_env_value,
+    check_config_version,
+    redact_key,
+)
+from gateway.status import get_running_pid, read_runtime_status
+
+try:
+    from fastapi import FastAPI, HTTPException
+    from fastapi.middleware.cors import CORSMiddleware
+    from fastapi.responses import FileResponse, JSONResponse
+    from fastapi.staticfiles import StaticFiles
+    from pydantic import BaseModel
+except ImportError:
+    raise SystemExit(
+        "Web UI requires fastapi and uvicorn.\n"
+        "Run 'hermes web' to auto-install, or: pip install hermes-agent[web]"
+    )
+
+WEB_DIST = Path(__file__).parent / "web_dist"
+
+app = FastAPI(title="Hermes Agent", version=__version__)
+
+# CORS: restrict to localhost origins only.  The web UI is intended to run
+# locally; binding to 0.0.0.0 with allow_origins=["*"] would let any website
+# read/modify config and secrets.
+_LOCALHOST_ORIGINS = [
+    "http://localhost:9119",
+    "http://127.0.0.1:9119",
+    "http://localhost:3000",   # Vite dev server
+    "http://127.0.0.1:3000",
+    "http://localhost:5173",   # Vite default
+    "http://127.0.0.1:5173",
+]
+
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=_LOCALHOST_ORIGINS,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+
+# ---------------------------------------------------------------------------
+# Config schema — auto-generated from DEFAULT_CONFIG
+# ---------------------------------------------------------------------------
+
+# Manual overrides for fields that need select options or custom types
+_SCHEMA_OVERRIDES: Dict[str, Dict[str, Any]] = {
+    "model": {
+        "type": "string",
+        "description": "Default model (e.g. anthropic/claude-sonnet-4.6)",
+    },
+    "terminal.backend": {
+        "type": "select",
+        "description": "Terminal execution backend",
+        "options": ["local", "docker", "ssh", "modal", "daytona", "singularity"],
+    },
+    "terminal.modal_mode": {
+        "type": "select",
+        "description": "Modal sandbox mode",
+        "options": ["sandbox", "function"],
+    },
+    "tts.provider": {
+        "type": "select",
+        "description": "Text-to-speech provider",
+        "options": ["edge", "elevenlabs", "openai", "neutts"],
+    },
+    "stt.provider": {
+        "type": "select",
+        "description": "Speech-to-text provider",
+        "options": ["local", "openai", "mistral"],
+    },
+    "display.skin": {
+        "type": "select",
+        "description": "CLI visual theme",
+        "options": ["default", "ares", "mono", "slate"],
+    },
+    "display.resume_display": {
+        "type": "select",
+        "description": "How resumed sessions display history",
+        "options": ["minimal", "full", "off"],
+    },
+    "display.busy_input_mode": {
+        "type": "select",
+        "description": "Input behavior while agent is running",
+        "options": ["queue", "interrupt", "block"],
+    },
+    "memory.provider": {
+        "type": "select",
+        "description": "Memory provider plugin",
+        "options": ["builtin", "honcho"],
+    },
+    "approvals.mode": {
+        "type": "select",
+        "description": "Dangerous command approval mode",
+        "options": ["ask", "yolo", "deny"],
+    },
+    "context.engine": {
+        "type": "select",
+        "description": "Context management engine",
+        "options": ["default", "custom"],
+    },
+    "human_delay.mode": {
+        "type": "select",
+        "description": "Simulated typing delay mode",
+        "options": ["off", "typing", "fixed"],
+    },
+    "logging.level": {
+        "type": "select",
+        "description": "Log level for agent.log",
+        "options": ["DEBUG", "INFO", "WARNING", "ERROR"],
+    },
+    "agent.service_tier": {
+        "type": "select",
+        "description": "API service tier (OpenAI/Anthropic)",
+        "options": ["", "auto", "default", "flex"],
+    },
+    "delegation.reasoning_effort": {
+        "type": "select",
+        "description": "Reasoning effort for delegated subagents",
+        "options": ["", "low", "medium", "high"],
+    },
+}
+
+
+def _infer_type(value: Any) -> str:
+    """Infer a UI field type from a Python value."""
+    if isinstance(value, bool):
+        return "boolean"
+    if isinstance(value, int):
+        return "number"
+    if isinstance(value, float):
+        return "number"
+    if isinstance(value, list):
+        return "list"
+    if isinstance(value, dict):
+        return "object"
+    return "string"
+
+
+def _build_schema_from_config(
+    config: Dict[str, Any],
+    prefix: str = "",
+) -> Dict[str, Dict[str, Any]]:
+    """Walk DEFAULT_CONFIG and produce a flat dot-path → field schema dict."""
+    schema: Dict[str, Dict[str, Any]] = {}
+    for key, value in config.items():
+        full_key = f"{prefix}{key}" if not prefix else f"{prefix}.{key}"
+        category = key if not prefix else prefix.split(".")[0]
+
+        # Skip internal / version keys
+        if full_key in ("_config_version",):
+            continue
+
+        if isinstance(value, dict):
+            # Recurse into nested dicts
+            schema.update(_build_schema_from_config(value, full_key))
+        else:
+            entry: Dict[str, Any] = {
+                "type": _infer_type(value),
+                "description": full_key.replace(".", " → ").replace("_", " ").title(),
+                "category": category,
+            }
+            # Apply manual overrides
+            if full_key in _SCHEMA_OVERRIDES:
+                entry.update(_SCHEMA_OVERRIDES[full_key])
+            schema[full_key] = entry
+    return schema
+
+
+CONFIG_SCHEMA = _build_schema_from_config(DEFAULT_CONFIG)
+
+
+class ConfigUpdate(BaseModel):
+    config: dict
+
+
+class EnvVarUpdate(BaseModel):
+    key: str
+    value: str
+
+
+class EnvVarDelete(BaseModel):
+    key: str
+
+
+@app.get("/api/status")
+async def get_status():
+    current_ver, latest_ver = check_config_version()
+
+    gateway_pid = get_running_pid()
+    gateway_running = gateway_pid is not None
+
+    gateway_state = None
+    gateway_platforms: dict = {}
+    gateway_exit_reason = None
+    gateway_updated_at = None
+    runtime = read_runtime_status()
+    if runtime:
+        gateway_state = runtime.get("gateway_state")
+        gateway_platforms = runtime.get("platforms") or {}
+        gateway_exit_reason = runtime.get("exit_reason")
+        gateway_updated_at = runtime.get("updated_at")
+        if not gateway_running:
+            gateway_state = gateway_state if gateway_state in ("stopped", "startup_failed") else "stopped"
+
+    active_sessions = 0
+    try:
+        from hermes_state import SessionDB
+        db = SessionDB()
+        sessions = db.list_sessions_rich(limit=50)
+        now = time.time()
+        active_sessions = sum(
+            1 for s in sessions
+            if s.get("ended_at") is None
+            and (now - s.get("last_active", s.get("started_at", 0))) < 300
+        )
+    except Exception:
+        pass
+
+    return {
+        "version": __version__,
+        "release_date": __release_date__,
+        "hermes_home": str(get_hermes_home()),
+        "config_path": str(get_config_path()),
+        "env_path": str(get_env_path()),
+        "config_version": current_ver,
+        "latest_config_version": latest_ver,
+        "gateway_running": gateway_running,
+        "gateway_pid": gateway_pid,
+        "gateway_state": gateway_state,
+        "gateway_platforms": gateway_platforms,
+        "gateway_exit_reason": gateway_exit_reason,
+        "gateway_updated_at": gateway_updated_at,
+        "active_sessions": active_sessions,
+    }
+
+
+@app.get("/api/sessions")
+async def get_sessions():
+    try:
+        from hermes_state import SessionDB
+        db = SessionDB()
+        sessions = db.list_sessions_rich(limit=20)
+        now = time.time()
+        for s in sessions:
+            s["is_active"] = (
+                s.get("ended_at") is None
+                and (now - s.get("last_active", s.get("started_at", 0))) < 300
+            )
+        return sessions
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))
+
+
+@app.get("/api/config")
+async def get_config():
+    return load_config()
+
+
+@app.get("/api/config/defaults")
+async def get_defaults():
+    return DEFAULT_CONFIG
+
+
+@app.get("/api/config/schema")
+async def get_schema():
+    return CONFIG_SCHEMA
+
+
+@app.put("/api/config")
+async def update_config(body: ConfigUpdate):
+    try:
+        save_config(body.config)
+        return {"ok": True}
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))
+
+
+@app.get("/api/env")
+async def get_env_vars():
+    env_on_disk = load_env()
+    result = {}
+    for var_name, info in OPTIONAL_ENV_VARS.items():
+        value = env_on_disk.get(var_name)
+        result[var_name] = {
+            "is_set": bool(value),
+            "redacted_value": redact_key(value) if value else None,
+            "description": info.get("description", ""),
+            "url": info.get("url"),
+            "category": info.get("category", ""),
+            "is_password": info.get("password", False),
+            "tools": info.get("tools", []),
+            "advanced": info.get("advanced", False),
+        }
+    return result
+
+
+@app.put("/api/env")
+async def set_env_var(body: EnvVarUpdate):
+    try:
+        save_env_value(body.key, body.value)
+        return {"ok": True, "key": body.key}
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))
+
+
+@app.delete("/api/env")
+async def remove_env_var(body: EnvVarDelete):
+    try:
+        removed = delete_env_value(body.key)
+        if not removed:
+            raise HTTPException(status_code=404, detail=f"{body.key} not found in .env")
+        return {"ok": True, "key": body.key}
+    except HTTPException:
+        raise
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))
+
+
+def mount_spa(application: FastAPI):
+    """Mount the built SPA. Falls back to index.html for client-side routing."""
+    if not WEB_DIST.exists():
+        @application.get("/{full_path:path}")
+        async def no_frontend(full_path: str):
+            return JSONResponse(
+                {"error": "Frontend not built. Run: cd web && npm run build"},
+                status_code=404,
+            )
+        return
+
+    application.mount("/assets", StaticFiles(directory=WEB_DIST / "assets"), name="assets")
+
+    @application.get("/{full_path:path}")
+    async def serve_spa(full_path: str):
+        file_path = WEB_DIST / full_path
+        if full_path and file_path.exists() and file_path.is_file():
+            return FileResponse(file_path)
+        return FileResponse(WEB_DIST / "index.html")
+
+
+mount_spa(app)
+
+
+def start_server(host: str = "127.0.0.1", port: int = 9119, open_browser: bool = True):
+    """Start the web UI server."""
+    import uvicorn
+
+    # Add the actual server origin to allowed CORS origins
+    server_origin = f"http://{host}:{port}"
+    if server_origin not in _LOCALHOST_ORIGINS:
+        _LOCALHOST_ORIGINS.append(server_origin)
+
+    if open_browser:
+        import threading
+        import webbrowser
+
+        def _open():
+            import time as _t
+            _t.sleep(1.0)
+            webbrowser.open(f"http://{host}:{port}")
+
+        threading.Thread(target=_open, daemon=True).start()
+
+    print(f"  Hermes Web UI → http://{host}:{port}")
+    uvicorn.run(app, host=host, port=port, log_level="warning")
@@ -189,78 +189,6 @@ def is_wsl() -> bool:
    return _wsl_detected


-# ─── Well-Known Paths ─────────────────────────────────────────────────────────
-
-
-def get_config_path() -> Path:
-    """Return the path to ``config.yaml`` under HERMES_HOME.
-
-    Replaces the ``get_hermes_home() / "config.yaml"`` pattern repeated
-    in 7+ files (skill_utils.py, hermes_logging.py, hermes_time.py, etc.).
-    """
-    return get_hermes_home() / "config.yaml"
-
-
-def get_skills_dir() -> Path:
-    """Return the path to the skills directory under HERMES_HOME."""
-    return get_hermes_home() / "skills"
-
-
-def get_logs_dir() -> Path:
-    """Return the path to the logs directory under HERMES_HOME."""
-    return get_hermes_home() / "logs"
-
-
-def get_env_path() -> Path:
-    """Return the path to the ``.env`` file under HERMES_HOME."""
-    return get_hermes_home() / ".env"
-
-
-# ─── Network Preferences ─────────────────────────────────────────────────────
-
-
-def apply_ipv4_preference(force: bool = False) -> None:
-    """Monkey-patch ``socket.getaddrinfo`` to prefer IPv4 connections.
-
-    On servers with broken or unreachable IPv6, Python tries AAAA records
-    first and hangs for the full TCP timeout before falling back to IPv4.
-    This affects httpx, requests, urllib, the OpenAI SDK — everything that
-    uses ``socket.getaddrinfo``.
-
-    When *force* is True, patches ``getaddrinfo`` so that calls with
-    ``family=AF_UNSPEC`` (the default) resolve as ``AF_INET`` instead,
-    skipping IPv6 entirely.  If no A record exists, falls back to the
-    original unfiltered resolution so pure-IPv6 hosts still work.
-
-    Safe to call multiple times — only patches once.
-    Set ``network.force_ipv4: true`` in ``config.yaml`` to enable.
-    """
-    if not force:
-        return
-
-    import socket
-
-    # Guard against double-patching
-    if getattr(socket.getaddrinfo, "_hermes_ipv4_patched", False):
-        return
-
-    _original_getaddrinfo = socket.getaddrinfo
-
-    def _ipv4_getaddrinfo(host, port, family=0, type=0, proto=0, flags=0):
-        if family == 0:  # AF_UNSPEC — caller didn't request a specific family
-            try:
-                return _original_getaddrinfo(
-                    host, port, socket.AF_INET, type, proto, flags
-                )
-            except socket.gaierror:
-                # No A record — fall back to full resolution (pure-IPv6 hosts)
-                return _original_getaddrinfo(host, port, family, type, proto, flags)
-        return _original_getaddrinfo(host, port, family, type, proto, flags)
-
-    _ipv4_getaddrinfo._hermes_ipv4_patched = True  # type: ignore[attr-defined]
-    socket.getaddrinfo = _ipv4_getaddrinfo  # type: ignore[assignment]
-
-
 OPENROUTER_BASE_URL = "https://openrouter.ai/api/v1"
 OPENROUTER_MODELS_URL = f"{OPENROUTER_BASE_URL}/models"

@@ -7,44 +7,27 @@ gateway call early in their startup path.  All log files live under
 Log files produced:
    agent.log   — INFO+, all agent/tool/session activity (the main log)
    errors.log  — WARNING+, errors and warnings only (quick triage)
-    gateway.log — INFO+, gateway-only events (created when mode="gateway")

-All files use ``RotatingFileHandler`` with ``RedactingFormatter`` so
+Both files use ``RotatingFileHandler`` with ``RedactingFormatter`` so
 secrets are never written to disk.
-
-Component separation:
-    gateway.log only receives records from ``gateway.*`` loggers —
-    platform adapters, session management, slash commands, delivery.
-    agent.log remains the catch-all (everything goes there).
-
-Session context:
-    Call ``set_session_context(session_id)`` at the start of a conversation
-    and ``clear_session_context()`` when done.  All log lines emitted on
-    that thread will include ``[session_id]`` for filtering/correlation.
 """

 import logging
 import os
-import threading
 from logging.handlers import RotatingFileHandler
 from pathlib import Path
-from typing import Optional, Sequence
+from typing import Optional

-from hermes_constants import get_config_path, get_hermes_home
+from hermes_constants import get_hermes_home

 # Sentinel to track whether setup_logging() has already run.  The function
 # is idempotent — calling it twice is safe but the second call is a no-op
 # unless ``force=True``.
 _logging_initialized = False

-# Thread-local storage for per-conversation session context.
-_session_context = threading.local()
-
-# Default log format — includes timestamp, level, optional session tag,
-# logger name, and message.  The ``%(session_tag)s`` field is guaranteed to
-# exist on every LogRecord via _install_session_record_factory() below.
-_LOG_FORMAT = "%(asctime)s %(levelname)s%(session_tag)s %(name)s: %(message)s"
-_LOG_FORMAT_VERBOSE = "%(asctime)s - %(name)s - %(levelname)s%(session_tag)s - %(message)s"
+# Default log format — includes timestamp, level, logger name, and message.
+_LOG_FORMAT = "%(asctime)s %(levelname)s %(name)s: %(message)s"
+_LOG_FORMAT_VERBOSE = "%(asctime)s - %(name)s - %(levelname)s - %(message)s"

 # Third-party loggers that are noisy at DEBUG/INFO level.
 _NOISY_LOGGERS = (
@@ -65,99 +48,6 @@ _NOISY_LOGGERS = (
 )


-# ---------------------------------------------------------------------------
-# Public session context API
-# ---------------------------------------------------------------------------
-
-def set_session_context(session_id: str) -> None:
-    """Set the session ID for the current thread.
-
-    All subsequent log records on this thread will include ``[session_id]``
-    in the formatted output.  Call at the start of ``run_conversation()``.
-    """
-    _session_context.session_id = session_id
-
-
-def clear_session_context() -> None:
-    """Clear the session ID for the current thread.
-
-    Optional — ``set_session_context()`` overwrites the previous value,
-    so explicit clearing is only needed if the thread is reused for
-    non-conversation work after ``run_conversation()`` returns.
-    """
-    _session_context.session_id = None
-
-
-# ---------------------------------------------------------------------------
-# Record factory — injects session_tag into every LogRecord at creation
-# ---------------------------------------------------------------------------
-
-def _install_session_record_factory() -> None:
-    """Replace the global LogRecord factory with one that adds ``session_tag``.
-
-    Unlike a ``logging.Filter`` on a handler or logger, the record factory
-    runs for EVERY record in the process — including records that propagate
-    from child loggers and records handled by third-party handlers.  This
-    guarantees ``%(session_tag)s`` is always available in format strings,
-    eliminating the KeyError that would occur if a handler used our format
-    without having a ``_SessionFilter`` attached.
-
-    Idempotent — checks for a marker attribute to avoid double-wrapping if
-    the module is reloaded.
-    """
-    current_factory = logging.getLogRecordFactory()
-    if getattr(current_factory, "_hermes_session_injector", False):
-        return  # already installed
-
-    def _session_record_factory(*args, **kwargs):
-        record = current_factory(*args, **kwargs)
-        sid = getattr(_session_context, "session_id", None)
-        record.session_tag = f" [{sid}]" if sid else ""  # type: ignore[attr-defined]
-        return record
-
-    _session_record_factory._hermes_session_injector = True  # type: ignore[attr-defined]
-    logging.setLogRecordFactory(_session_record_factory)
-
-
-# Install immediately on import — session_tag is available on all records
-# from this point forward, even before setup_logging() is called.
-_install_session_record_factory()
-
-
-# ---------------------------------------------------------------------------
-# Filters
-# ---------------------------------------------------------------------------
-
-class _ComponentFilter(logging.Filter):
-    """Only pass records whose logger name starts with one of *prefixes*.
-
-    Used to route gateway-specific records to ``gateway.log`` while
-    keeping ``agent.log`` as the catch-all.
-    """
-
-    def __init__(self, prefixes: Sequence[str]) -> None:
-        super().__init__()
-        self._prefixes = tuple(prefixes)
-
-    def filter(self, record: logging.LogRecord) -> bool:
-        return record.name.startswith(self._prefixes)
-
-
-# Logger name prefixes that belong to each component.
-# Used by _ComponentFilter and exposed for ``hermes logs --component``.
-COMPONENT_PREFIXES = {
-    "gateway": ("gateway",),
-    "agent": ("agent", "run_agent", "model_tools", "batch_runner"),
-    "tools": ("tools",),
-    "cli": ("hermes_cli", "cli"),
-    "cron": ("cron",),
-}
-
-
-# ---------------------------------------------------------------------------
-# Main setup
-# ---------------------------------------------------------------------------
-
 def setup_logging(
    *,
    hermes_home: Optional[Path] = None,
@@ -188,9 +78,8 @@ def setup_logging(
        Number of rotated backup files to keep.
        Defaults to 3 or the value from config.yaml ``logging.backup_count``.
    mode
-        Caller context: ``"cli"``, ``"gateway"``, ``"cron"``.
-        When ``"gateway"``, an additional ``gateway.log`` file is created
-        that receives only gateway-component records.
+        Hint for the caller context: ``"cli"``, ``"gateway"``, ``"cron"``.
+        Currently used only for log format tuning (gateway includes PID).
    force
        Re-run setup even if it has already been called.

@@ -241,18 +130,6 @@ def setup_logging(
        formatter=RedactingFormatter(_LOG_FORMAT),
    )

-    # --- gateway.log (INFO+, gateway component only) ------------------------
-    if mode == "gateway":
-        _add_rotating_handler(
-            root,
-            log_dir / "gateway.log",
-            level=logging.INFO,
-            max_bytes=5 * 1024 * 1024,
-            backup_count=3,
-            formatter=RedactingFormatter(_LOG_FORMAT),
-            log_filter=_ComponentFilter(COMPONENT_PREFIXES["gateway"]),
-        )
-
    # Ensure root logger level is low enough for the handlers to fire.
    if root.level == logging.NOTSET or root.level > level:
        root.setLevel(level)
@@ -341,16 +218,9 @@ def _add_rotating_handler(
    max_bytes: int,
    backup_count: int,
    formatter: logging.Formatter,
-    log_filter: Optional[logging.Filter] = None,
 ) -> None:
    """Add a ``RotatingFileHandler`` to *logger*, skipping if one already
    exists for the same resolved file path (idempotent).
-
-    Parameters
-    ----------
-    log_filter
-        Optional filter to attach to the handler (e.g. ``_ComponentFilter``
-        for gateway.log).
    """
    resolved = path.resolve()
    for existing in logger.handlers:
@@ -366,8 +236,6 @@ def _add_rotating_handler(
    )
    handler.setLevel(level)
    handler.setFormatter(formatter)
-    if log_filter is not None:
-        handler.addFilter(log_filter)
    logger.addHandler(handler)


@@ -378,7 +246,7 @@ def _read_logging_config():
    """
    try:
        import yaml
-        config_path = get_config_path()
+        config_path = get_hermes_home() / "config.yaml"
        if config_path.exists():
            with open(config_path, "r", encoding="utf-8") as f:
                cfg = yaml.safe_load(f) or {}
@@ -16,7 +16,7 @@ crashes due to a bad timezone string.
 import logging
 import os
 from datetime import datetime
-from hermes_constants import get_config_path
+from hermes_constants import get_hermes_home
 from typing import Optional

 logger = logging.getLogger(__name__)
@@ -48,7 +48,8 @@ def _resolve_timezone_name() -> str:
    # 2. config.yaml ``timezone`` key
    try:
        import yaml
-        config_path = get_config_path()
+        hermes_home = get_hermes_home()
+        config_path = hermes_home / "config.yaml"
        if config_path.exists():
            with open(config_path) as f:
                cfg = yaml.safe_load(f) or {}
@@ -499,16 +499,6 @@
          default = "ubuntu:24.04";
          description = "OCI container image. The container pulls this at runtime via Docker/Podman.";
        };
-
-        hostUsers = mkOption {
-          type = types.listOf types.str;
-          default = [ ];
-          description = ''
-            Interactive users who get a ~/.hermes symlink to the service
-            stateDir. These users are automatically added to the hermes group.
-          '';
-          example = [ "sidbin" ];
-        };
      };
    };

@@ -567,25 +557,6 @@
        environment.variables.HERMES_HOME = "${cfg.stateDir}/.hermes";
      })

-      # ── Host user group membership ─────────────────────────────────────
-      (lib.mkIf (cfg.container.enable && cfg.container.hostUsers != []) {
-        users.users = lib.genAttrs cfg.container.hostUsers (user: {
-          extraGroups = [ cfg.group ];
-        });
-      })
-
-      # ── Warnings ──────────────────────────────────────────────────────
-      (lib.mkIf (cfg.container.enable && !cfg.addToSystemPackages && cfg.container.hostUsers != []) {
-        warnings = [
-          ''
-            services.hermes-agent: container.enable is true and container.hostUsers
-            is set, but addToSystemPackages is false. Without a host-installed hermes
-            binary, container routing will not work for interactive users.
-            Set addToSystemPackages = true or ensure hermes is on PATH.
-          ''
-        ];
-      })
-
      # ── Directories ───────────────────────────────────────────────────
      {
        systemd.tmpfiles.rules = [
@@ -640,59 +611,6 @@
          chown ${cfg.user}:${cfg.group} ${cfg.stateDir}/.hermes/.managed
          chmod 0644 ${cfg.stateDir}/.hermes/.managed

-          # Container mode metadata — tells the host CLI to exec into the
-          # container instead of running locally. Removed when container mode
-          # is disabled so the host CLI falls back to native execution.
-          ${if cfg.container.enable then ''
-            cat > ${cfg.stateDir}/.hermes/.container-mode <<'HERMES_CONTAINER_MODE_EOF'
-# Written by NixOS activation script. Do not edit manually.
-backend=${cfg.container.backend}
-container_name=${containerName}
-exec_user=${cfg.user}
-hermes_bin=${containerDataDir}/current-package/bin/hermes
-HERMES_CONTAINER_MODE_EOF
-            chown ${cfg.user}:${cfg.group} ${cfg.stateDir}/.hermes/.container-mode
-            chmod 0644 ${cfg.stateDir}/.hermes/.container-mode
-          '' else ''
-            rm -f ${cfg.stateDir}/.hermes/.container-mode
-
-            # Remove symlink bridge for hostUsers
-            ${lib.concatStringsSep "\n" (map (user:
-              let
-                userHome = config.users.users.${user}.home;
-                symlinkPath = "${userHome}/.hermes";
-              in ''
-                if [ -L "${symlinkPath}" ] && [ "$(readlink "${symlinkPath}")" = "${cfg.stateDir}/.hermes" ]; then
-                  rm -f "${symlinkPath}"
-                  echo "hermes-agent: removed symlink ${symlinkPath}"
-                fi
-              '') cfg.container.hostUsers)}
-          ''}
-
-          # ── Symlink bridge for interactive users ───────────────────────
-          # Create ~/.hermes -> stateDir/.hermes for each hostUser so the
-          # host CLI shares state with the container service.
-          # Only runs when container mode is enabled.
-          ${lib.optionalString cfg.container.enable
-            (lib.concatStringsSep "\n" (map (user:
-              let
-                userHome = config.users.users.${user}.home;
-                symlinkPath = "${userHome}/.hermes";
-                target = "${cfg.stateDir}/.hermes";
-              in ''
-                if [ -d "${symlinkPath}" ] && [ ! -L "${symlinkPath}" ]; then
-                  # Real directory — back it up, then create symlink.
-                  # (ln -sfn cannot atomically replace a directory.)
-                  _backup="${symlinkPath}.bak.$(date +%s)"
-                  echo "hermes-agent: backing up existing ${symlinkPath} to $_backup"
-                  mv "${symlinkPath}" "$_backup"
-                fi
-                # For everything else (existing symlink, doesn't exist, etc.)
-                # ln -sfn handles it: replaces symlinks, creates new ones.
-                ln -sfn "${target}" "${symlinkPath}"
-                chown -h ${user}:${cfg.group} "${symlinkPath}"
-              '') cfg.container.hostUsers))}
-
          # Seed auth file if provided
          ${lib.optionalString (cfg.authFile != null) ''
            ${if cfg.authFileForceOverwrite then ''
@@ -376,24 +376,6 @@ def backup_existing(path: Path, backup_root: Path) -> Optional[Path]:
    return dest


-# ── Brand rewriting ─────────────────────────────────────────
-# Replace OpenClaw brand names with Hermes in migrated text so that
-# memory entries, user profiles, SOUL.md, and workspace instructions
-# read as self-referential to the new agent identity.
-_REBRAND_PATTERNS: List[Tuple[re.Pattern, str]] = [
-    (re.compile(r'\bOpen[\s-]?Claw\b', re.IGNORECASE), 'Hermes'),
-    (re.compile(r'\bClawdBot\b', re.IGNORECASE), 'Hermes'),
-    (re.compile(r'\bMoltBot\b', re.IGNORECASE), 'Hermes'),
-]
-
-
-def rebrand_text(text: str) -> str:
-    """Replace OpenClaw / ClawdBot / MoltBot brand names with Hermes."""
-    for pattern, replacement in _REBRAND_PATTERNS:
-        text = pattern.sub(replacement, text)
-    return text
-
-
 def parse_existing_memory_entries(path: Path) -> List[str]:
    if not path.exists():
        return []
@@ -635,19 +617,6 @@ class Migrator:
            candidate = self.source_root / rel
            if candidate.exists():
                return candidate
-            # OpenClaw renamed workspace/ to workspace-main/ (and workspace-{agentId}
-            # for multi-agent).  Try the new path as a fallback.
-            if rel.startswith("workspace/"):
-                suffix = rel[len("workspace/"):]
-                for variant in ("workspace-main", "workspace-assistant"):
-                    alt = self.source_root / variant / suffix
-                    if alt.exists():
-                        return alt
-            elif rel.startswith("workspace.default/"):
-                suffix = rel[len("workspace.default/"):]
-                alt = self.source_root / "workspace-main" / suffix
-                if alt.exists():
-                    return alt
        return None

    def resolve_skill_destination(self, destination: Path) -> Path:
@@ -800,13 +769,12 @@ class Migrator:
        path.write_text("\n".join(entries) + "\n", encoding="utf-8")
        return path

-    def copy_file(self, source: Path, destination: Path, kind: str,
-                  transform: Optional[Any] = None) -> None:
+    def copy_file(self, source: Path, destination: Path, kind: str) -> None:
        if not source or not source.exists():
            return

        if destination.exists():
-            if not transform and sha256_file(source) == sha256_file(destination):
+            if sha256_file(source) == sha256_file(destination):
                self.record(kind, source, destination, "skipped", "Target already matches source")
                return
            if not self.overwrite:
@@ -816,13 +784,7 @@ class Migrator:
        if self.execute:
            backup_path = self.maybe_backup(destination)
            ensure_parent(destination)
-            if transform:
-                content = read_text(source)
-                content = transform(content)
-                destination.write_text(content, encoding="utf-8")
-                shutil.copystat(source, destination)
-            else:
-                shutil.copy2(source, destination)
+            shutil.copy2(source, destination)
            self.record(kind, source, destination, "migrated", backup=str(backup_path) if backup_path else None)
        else:
            self.record(kind, source, destination, "migrated", "Would copy")
@@ -832,7 +794,7 @@ class Migrator:
        if not source:
            self.record("soul", None, self.target_root / "SOUL.md", "skipped", "No OpenClaw SOUL.md found")
            return
-        self.copy_file(source, self.target_root / "SOUL.md", kind="soul", transform=rebrand_text)
+        self.copy_file(source, self.target_root / "SOUL.md", kind="soul")

    def migrate_workspace_agents(self) -> None:
        source = self.source_candidate(
@@ -846,7 +808,7 @@ class Migrator:
            self.record("workspace-agents", source, None, "skipped", "No workspace target was provided")
            return
        destination = self.workspace_target / WORKSPACE_INSTRUCTIONS_FILENAME
-        self.copy_file(source, destination, kind="workspace-agents", transform=rebrand_text)
+        self.copy_file(source, destination, kind="workspace-agents")

    def migrate_memory(self, source: Optional[Path], destination: Path, limit: int, kind: str) -> None:
        if not source or not source.exists():
@@ -857,7 +819,6 @@ class Migrator:
        if not incoming:
            self.record(kind, source, destination, "skipped", "No importable entries found")
            return
-        incoming = [rebrand_text(entry) for entry in incoming]

        existing = parse_existing_memory_entries(destination)
        merged, stats, overflowed = merge_entries(existing, incoming, limit)
@@ -953,7 +914,7 @@ class Migrator:

    def load_openclaw_config(self) -> Dict[str, Any]:
        # Check current name and legacy config filenames
-        for name in ("openclaw.json", "clawdbot.json", "moltbot.json"):
+        for name in ("openclaw.json", "clawdbot.json", "moldbot.json"):
            config_path = self.source_root / name
            if config_path.exists():
                try:
@@ -1023,17 +984,7 @@ class Migrator:
            .get("workspace")
        )
        if isinstance(workspace, str) and workspace.strip():
-            ws_path = workspace.strip()
-            # Skip if the workspace points inside the OpenClaw source directory —
-            # that path will be stale after migration and would cause the Hermes
-            # gateway to use the old OpenClaw workspace as its cwd, picking up
-            # OpenClaw's AGENTS.md, MEMORY.md, etc.
-            try:
-                inside_source = Path(ws_path).resolve().is_relative_to(self.source_root.resolve())
-            except (ValueError, OSError):
-                inside_source = False
-            if not inside_source:
-                additions["MESSAGING_CWD"] = ws_path
+            additions["MESSAGING_CWD"] = workspace.strip()

        allowlist_path = self.source_root / "credentials" / "telegram-default-allowFrom.json"
        if allowlist_path.exists():
@@ -1082,8 +1033,11 @@ class Migrator:
    def migrate_secret_settings(self, config: Dict[str, Any]) -> None:
        secret_additions: Dict[str, str] = {}

-        tg_cfg = config.get("channels", {}).get("telegram", {})
-        telegram_token = self._get_channel_field(tg_cfg, "botToken") if isinstance(tg_cfg, dict) else None
+        telegram_token = (
+            config.get("channels", {})
+            .get("telegram", {})
+            .get("botToken")
+        )
        if isinstance(telegram_token, str) and telegram_token.strip():
            secret_additions["TELEGRAM_BOT_TOKEN"] = telegram_token.strip()

@@ -1103,28 +1057,15 @@ class Migrator:
        """Resolve a channel config value that may be a SecretRef."""
        return resolve_secret_input(value, self.load_openclaw_env())

-    @staticmethod
-    def _get_channel_field(ch_cfg: Dict[str, Any], field: str) -> Any:
-        """Get a field from channel config, checking both flat and accounts.default layout."""
-        val = ch_cfg.get(field)
-        if val is not None:
-            return val
-        accounts = ch_cfg.get("accounts")
-        if isinstance(accounts, dict):
-            default = accounts.get("default")
-            if isinstance(default, dict):
-                return default.get(field)
-        return None
-
    def migrate_discord_settings(self, config: Optional[Dict[str, Any]] = None) -> None:
        config = config or self.load_openclaw_config()
        additions: Dict[str, str] = {}
        discord = config.get("channels", {}).get("discord", {})
        if isinstance(discord, dict):
-            token = self._get_channel_field(discord, "token")
+            token = discord.get("token")
            if isinstance(token, str) and token.strip():
                additions["DISCORD_BOT_TOKEN"] = token.strip()
-            allow_from = self._get_channel_field(discord, "allowFrom") or []
+            allow_from = discord.get("allowFrom", [])
            if isinstance(allow_from, list):
                users = [str(u).strip() for u in allow_from if str(u).strip()]
                if users:
@@ -1139,13 +1080,13 @@ class Migrator:
        additions: Dict[str, str] = {}
        slack = config.get("channels", {}).get("slack", {})
        if isinstance(slack, dict):
-            bot_token = self._get_channel_field(slack, "botToken")
+            bot_token = slack.get("botToken")
            if isinstance(bot_token, str) and bot_token.strip():
                additions["SLACK_BOT_TOKEN"] = bot_token.strip()
-            app_token = self._get_channel_field(slack, "appToken")
+            app_token = slack.get("appToken")
            if isinstance(app_token, str) and app_token.strip():
                additions["SLACK_APP_TOKEN"] = app_token.strip()
-            allow_from = self._get_channel_field(slack, "allowFrom") or []
+            allow_from = slack.get("allowFrom", [])
            if isinstance(allow_from, list):
                users = [str(u).strip() for u in allow_from if str(u).strip()]
                if users:
@@ -1160,7 +1101,7 @@ class Migrator:
        additions: Dict[str, str] = {}
        whatsapp = config.get("channels", {}).get("whatsapp", {})
        if isinstance(whatsapp, dict):
-            allow_from = self._get_channel_field(whatsapp, "allowFrom") or []
+            allow_from = whatsapp.get("allowFrom", [])
            if isinstance(allow_from, list):
                users = [str(u).strip() for u in allow_from if str(u).strip()]
                if users:
@@ -1175,13 +1116,13 @@ class Migrator:
        additions: Dict[str, str] = {}
        signal = config.get("channels", {}).get("signal", {})
        if isinstance(signal, dict):
-            account = self._get_channel_field(signal, "account")
+            account = signal.get("account")
            if isinstance(account, str) and account.strip():
                additions["SIGNAL_ACCOUNT"] = account.strip()
-            http_url = self._get_channel_field(signal, "httpUrl")
+            http_url = signal.get("httpUrl")
            if isinstance(http_url, str) and http_url.strip():
                additions["SIGNAL_HTTP_URL"] = http_url.strip()
-            allow_from = self._get_channel_field(signal, "allowFrom") or []
+            allow_from = signal.get("allowFrom", [])
            if isinstance(allow_from, list):
                users = [str(u).strip() for u in allow_from if str(u).strip()]
                if users:
@@ -1220,16 +1161,6 @@ class Migrator:
                raw_key = provider_cfg.get("apiKey")
                api_key = resolve_secret_input(raw_key, openclaw_env)
                if not api_key:
-                    # Warn if a SecretRef with file/exec source was silently unresolvable
-                    if isinstance(raw_key, dict) and raw_key.get("source") in ("file", "exec"):
-                        self.record(
-                            "provider-keys",
-                            self.source_root / "openclaw.json",
-                            None,
-                            "skipped",
-                            f"Provider '{provider_name}' uses a {raw_key['source']}-backed SecretRef "
-                            f"that cannot be auto-migrated. Add this key manually via: hermes config set",
-                        )
                    continue

                base_url = provider_cfg.get("baseUrl", "")
@@ -1293,21 +1224,6 @@ class Migrator:
            if val and hermes_key not in secret_additions:
                secret_additions[hermes_key] = val

-        # Check the openclaw.json "env" sub-object — some OpenClaw setups
-        # store API keys here instead of in a separate .env file.
-        # Keys can be at env.<KEY> or env.vars.<KEY>.
-        json_env = config.get("env")
-        if isinstance(json_env, dict):
-            env_vars = json_env.get("vars")
-            sources = [json_env]
-            if isinstance(env_vars, dict):
-                sources.append(env_vars)
-            for src in sources:
-                for oc_key, hermes_key in env_key_mapping.items():
-                    val = src.get(oc_key)
-                    if isinstance(val, str) and val.strip() and hermes_key not in secret_additions:
-                        secret_additions[hermes_key] = val.strip()
-
        # Check per-agent auth-profiles.json for additional credentials
        auth_profiles_path = self.source_root / "agents" / "main" / "agent" / "auth-profiles.json"
        if auth_profiles_path.exists():
@@ -1408,9 +1324,8 @@ class Migrator:
        tts_data: Dict[str, Any] = {}

        provider = tts.get("provider")
-        if isinstance(provider, str) and provider in ("elevenlabs", "openai", "edge", "microsoft"):
-            # OpenClaw renamed "edge" to "microsoft"; Hermes still uses "edge"
-            tts_data["provider"] = "edge" if provider == "microsoft" else provider
+        if isinstance(provider, str) and provider in ("elevenlabs", "openai", "edge"):
+            tts_data["provider"] = provider

        # TTS provider settings live under messages.tts.providers.{provider}
        # in OpenClaw (not messages.tts.elevenlabs directly)
@@ -1459,9 +1374,9 @@ class Migrator:
                tts_data["openai"] = oai_settings

        edge_tts = (
-            (providers.get("edge") or providers.get("microsoft") or {})
-            if isinstance(providers.get("edge"), dict) or isinstance(providers.get("microsoft"), dict) else
-            (tts.get("edge") or tts.get("microsoft") or {})
+            (providers.get("edge") or {})
+            if isinstance(providers.get("edge"), dict) else
+            (tts.get("edge") or {})
        )
        if isinstance(edge_tts, dict):
            edge_voice = edge_tts.get("voice")
@@ -1579,7 +1494,6 @@ class Migrator:
        if not all_incoming:
            self.record("daily-memory", source_dir, destination, "skipped", "No importable entries found in daily memory files")
            return
-        all_incoming = [rebrand_text(entry) for entry in all_incoming]

        existing = parse_existing_memory_entries(destination)
        merged, stats, overflowed = merge_entries(existing, all_incoming, self.memory_limit)
@@ -1976,11 +1890,11 @@ class Migrator:
        if defaults.get("thinkingDefault"):
            # Map OpenClaw thinking -> Hermes reasoning_effort
            thinking = defaults["thinkingDefault"]
-            if thinking in ("always", "high", "xhigh"):
+            if thinking in ("always", "high"):
                agent_cfg["reasoning_effort"] = "high"
-            elif thinking in ("auto", "medium", "adaptive"):
+            elif thinking in ("auto", "medium"):
                agent_cfg["reasoning_effort"] = "medium"
-            elif thinking in ("off", "low", "none", "minimal"):
+            elif thinking in ("off", "low", "none"):
                agent_cfg["reasoning_effort"] = "low"
            changes = True

@@ -2185,14 +2099,10 @@ class Migrator:
                                f"Provider '{prov_name}' already exists")
                    continue

-                api_type = prov_cfg.get("apiType") or prov_cfg.get("api") or prov_cfg.get("type") or "openai"
+                api_type = prov_cfg.get("apiType") or prov_cfg.get("type") or "openai"
                api_mode_map = {
                    "openai": "chat_completions",
-                    "openai-completions": "chat_completions",
-                    "openai-responses": "chat_completions",
                    "anthropic": "anthropic_messages",
-                    "anthropic-messages": "anthropic_messages",
-                    "google-generative-ai": "chat_completions",
                    "cohere": "chat_completions",
                }
                entry = {
@@ -2232,7 +2142,7 @@ class Migrator:

        # Extended channel token/allowlist mapping
        CHANNEL_ENV_MAP = {
-            "matrix": {"token": "MATRIX...OKEN", "tokenField": "accessToken", "allowFrom": "MATRIX_ALLOWED_USERS",
+            "matrix": {"token": "MATRIX_ACCESS_TOKEN", "allowFrom": "MATRIX_ALLOWED_USERS",
                        "extras": {"homeserverUrl": "MATRIX_HOMESERVER_URL", "userId": "MATRIX_USER_ID"}},
            "mattermost": {"token": "MATTERMOST_BOT_TOKEN", "allowFrom": "MATTERMOST_ALLOWED_USERS",
                           "extras": {"url": "MATTERMOST_URL", "teamId": "MATTERMOST_TEAM_ID"}},
@@ -2250,21 +2160,19 @@ class Migrator:
            if not ch_cfg:
                continue

-            # Extract tokens (check flat path, then accounts.default)
-            token_field = ch_mapping.get("tokenField", "botToken")
-            bot_token = self._get_channel_field(ch_cfg, token_field)
-            if ch_mapping.get("token") and bot_token and self.migrate_secrets:
-                self._set_env_var(ch_mapping["token"], str(bot_token),
-                                  f"channels.{ch_name}.{token_field}")
-            allow_val = self._get_channel_field(ch_cfg, "allowFrom")
-            if ch_mapping.get("allowFrom") and allow_val:
+            # Extract tokens
+            if ch_mapping.get("token") and ch_cfg.get("botToken") and self.migrate_secrets:
+                self._set_env_var(ch_mapping["token"], ch_cfg["botToken"],
+                                  f"channels.{ch_name}.botToken")
+            if ch_mapping.get("allowFrom") and ch_cfg.get("allowFrom"):
+                allow_val = ch_cfg["allowFrom"]
                if isinstance(allow_val, list):
                    allow_val = ",".join(str(x) for x in allow_val)
                self._set_env_var(ch_mapping["allowFrom"], str(allow_val),
                                  f"channels.{ch_name}.allowFrom")
            # Extra fields
            for oc_key, env_key in (ch_mapping.get("extras") or {}).items():
-                val = self._get_channel_field(ch_cfg, oc_key)
+                val = ch_cfg.get(oc_key)
                if val:
                    if isinstance(val, list):
                        val = ",".join(str(x) for x in val)
@@ -2587,33 +2495,6 @@ class Migrator:
        elif has_cron_store_archive:
            notes.append("- Run `hermes cron` to recreate scheduled tasks (see archived cron-store)")

-        # Check if skills were imported
-        has_skills = any(i.kind == "skills" and i.status == "migrated" for i in self.items)
-        if has_skills:
-            notes.extend([
-                "",
-                "## Imported Skills",
-                "",
-                "Imported skills require a new session to take effect. After migration,",
-                "restart your agent or start a new chat session, then run `/skills`",
-                "to verify they loaded correctly.",
-                "",
-            ])
-
-        # Check if WhatsApp was detected
-        has_whatsapp = any(i.kind == "whatsapp-settings" and i.status == "migrated" for i in self.items)
-        if has_whatsapp:
-            notes.extend([
-                "",
-                "## WhatsApp Requires Re-Pairing",
-                "",
-                "WhatsApp uses QR-code pairing, not token-based auth. Your allowlist",
-                "was migrated, but you must re-pair the device by running:",
-                "",
-                "    hermes whatsapp",
-                "",
-            ])
-
        notes.extend([
            "- Run `hermes gateway install` if you need the gateway service",
            "- Review `~/.hermes/config.yaml` for any adjustments",
@@ -43,7 +43,7 @@ dev = ["debugpy>=1.8.0,<2", "pytest>=9.0.2,<10", "pytest-asyncio>=1.3.0,<2", "py
 messaging = ["python-telegram-bot[webhooks]>=22.6,<23", "discord.py[voice]>=2.7.1,<3", "aiohttp>=3.13.3,<4", "slack-bolt>=1.18.0,<2", "slack-sdk>=3.27.0,<4"]
 cron = ["croniter>=6.0.0,<7"]
 slack = ["slack-bolt>=1.18.0,<2", "slack-sdk>=3.27.0,<4"]
-matrix = ["mautrix[encryption]>=0.20,<1", "Markdown>=3.6,<4", "aiosqlite>=0.20", "asyncpg>=0.29"]
+matrix = ["mautrix[encryption]>=0.20,<1", "Markdown>=3.6,<4"]
 cli = ["simple-term-menu>=1.0,<2"]
 tts-premium = ["elevenlabs>=1.0,<2"]
 voice = [
@@ -76,6 +76,7 @@ termux = [
 ]
 dingtalk = ["dingtalk-stream>=0.1.0,<1"]
 feishu = ["lark-oapi>=1.5.3,<2"]
+web = ["fastapi>=0.104.0,<1", "uvicorn[standard]>=0.24.0,<1"]
 rl = [
  "atroposlib @ git+https://github.com/NousResearch/atropos.git",
  "tinker @ git+https://github.com/thinking-machines-lab/tinker.git",
@@ -107,6 +108,7 @@ all = [
  "hermes-agent[dingtalk]",
  "hermes-agent[feishu]",
  "hermes-agent[mistral]",
+  "hermes-agent[web]",
 ]

 [project.scripts]
@@ -117,6 +119,9 @@ hermes-acp = "acp_adapter.entry:main"
 [tool.setuptools]
 py-modules = ["run_agent", "model_tools", "toolsets", "batch_runner", "trajectory_compressor", "toolset_distributions", "cli", "hermes_constants", "hermes_state", "hermes_time", "hermes_logging", "rl_cli", "utils"]

+[tool.setuptools.package-data]
+hermes_cli = ["web_dist/**/*"]
+
 [tool.setuptools.packages.find]
 include = ["agent", "tools", "tools.*", "hermes_cli", "gateway", "gateway.*", "cron", "acp_adapter", "plugins", "plugins.*"]

@@ -339,7 +339,10 @@ def _paths_overlap(left: Path, right: Path) -> bool:

 _SURROGATE_RE = re.compile(r'[\ud800-\udfff]')

-
+_BUDGET_WARNING_RE = re.compile(
+    r"\[BUDGET(?:\s+WARNING)?:\s+Iteration\s+\d+/\d+\..*?\]",
+    re.DOTALL,
+)


 def _sanitize_surrogates(text: str) -> str:
@@ -460,7 +463,34 @@ def _sanitize_messages_non_ascii(messages: list) -> bool:
    return found


+def _strip_budget_warnings_from_history(messages: list) -> None:
+    """Remove budget pressure warnings from tool-result messages in-place.

+    Budget warnings are turn-scoped signals that must not leak into replayed
+    history.  They live in tool-result ``content`` either as a JSON key
+    (``_budget_warning``) or appended plain text.
+    """
+    for msg in messages:
+        if not isinstance(msg, dict) or msg.get("role") != "tool":
+            continue
+        content = msg.get("content")
+        if not isinstance(content, str) or "_budget_warning" not in content and "[BUDGET" not in content:
+            continue
+
+        # Try JSON first (the common case: _budget_warning key in a dict)
+        try:
+            parsed = json.loads(content)
+            if isinstance(parsed, dict) and "_budget_warning" in parsed:
+                del parsed["_budget_warning"]
+                msg["content"] = json.dumps(parsed, ensure_ascii=False)
+                continue
+        except (json.JSONDecodeError, TypeError):
+            pass
+
+        # Fallback: strip the text pattern from plain-text tool results
+        cleaned = _BUDGET_WARNING_RE.sub("", content).strip()
+        if cleaned != content:
+            msg["content"] = cleaned


 # =========================================================================
@@ -549,7 +579,6 @@ class AIAgent:
        clarify_callback: callable = None,
        step_callback: callable = None,
        stream_delta_callback: callable = None,
-        interim_assistant_callback: callable = None,
        tool_gen_callback: callable = None,
        status_callback: callable = None,
        max_tokens: int = None,
@@ -671,14 +700,10 @@ class AIAgent:
        except Exception:
            pass

-        # GPT-5.x models require the Responses API path — they are rejected
-        # on /v1/chat/completions by both OpenAI and OpenRouter.  Also
-        # auto-upgrade for direct OpenAI URLs (api.openai.com) since all
-        # newer tool-calling models prefer Responses there.
-        if self.api_mode == "chat_completions" and (
-            self._is_direct_openai_url()
-            or self._model_requires_responses_api(self.model)
-        ):
+        # Direct OpenAI sessions use the Responses API path.  GPT-5.x tool
+        # calls with reasoning are rejected on /v1/chat/completions, and
+        # Hermes is a tool-using client by default.
+        if self.api_mode == "chat_completions" and self._is_direct_openai_url():
            self.api_mode = "codex_responses"

        # Pre-warm OpenRouter model metadata cache in a background thread.
@@ -699,7 +724,6 @@ class AIAgent:
        self.clarify_callback = clarify_callback
        self.step_callback = step_callback
        self.stream_delta_callback = stream_delta_callback
-        self.interim_assistant_callback = interim_assistant_callback
        self.status_callback = status_callback
        self.tool_gen_callback = tool_gen_callback

@@ -711,7 +735,6 @@ class AIAgent:
        # Interrupt mechanism for breaking out of tool loops
        self._interrupt_requested = False
        self._interrupt_message = None  # Optional message that triggered interrupt
-        self._execution_thread_id: int | None = None  # Set at run_conversation() start
        self._client_lock = threading.RLock()
        
        # Subagent delegation state
@@ -743,18 +766,16 @@ class AIAgent:
        # conversation prefix. Uses system_and_3 strategy (4 breakpoints).
        is_openrouter = self._is_openrouter_url()
        is_claude = "claude" in self.model.lower()
-        is_native_anthropic = self.api_mode == "anthropic_messages" and self.provider == "anthropic"
+        is_native_anthropic = self.api_mode == "anthropic_messages"
        self._use_prompt_caching = (is_openrouter and is_claude) or is_native_anthropic
        self._cache_ttl = "5m"  # Default 5-minute TTL (1.25x write cost)
        
-        # Iteration budget: the LLM is only notified when it actually exhausts
-        # the iteration budget (api_call_count >= max_iterations).  At that
-        # point we inject ONE message, allow one final API call, and if the
-        # model doesn't produce a text response, force a user-message asking
-        # it to summarise.  No intermediate pressure warnings — they caused
-        # models to "give up" prematurely on complex tasks (#7915).
-        self._budget_exhausted_injected = False
-        self._budget_grace_call = False
+        # Iteration budget pressure: warn the LLM as it approaches max_iterations.
+        # Warnings are injected into the last tool result JSON (not as separate
+        # messages) so they don't break message structure or invalidate caching.
+        self._budget_caution_threshold = 0.7   # 70% — nudge to start wrapping up
+        self._budget_warning_threshold = 0.9   # 90% — urgent, respond now
+        self._budget_pressure_enabled = True

        # Context pressure warnings: notify the USER (not the LLM) as context
        # fills up.  Purely informational — displayed in CLI output and sent via
@@ -805,11 +826,6 @@ class AIAgent:
        # Deferred paragraph break flag — set after tool iterations so a
        # single "\n\n" is prepended to the next real text delta.
        self._stream_needs_break = False
-        # Visible assistant text already delivered through live token callbacks
-        # during the current model response. Used to avoid re-sending the same
-        # commentary when the provider later returns it as a completed interim
-        # assistant message.
-        self._current_streamed_assistant_text = ""

        # Optional current-turn user-message override used when the API-facing
        # user message intentionally differs from the persisted transcript
@@ -1310,19 +1326,6 @@ class AIAgent:
            )
        self.compression_enabled = compression_enabled

-        # Reject models whose context window is below the minimum required
-        # for reliable tool-calling workflows (64K tokens).
-        from agent.model_metadata import MINIMUM_CONTEXT_LENGTH
-        _ctx = getattr(self.context_compressor, "context_length", 0)
-        if _ctx and _ctx < MINIMUM_CONTEXT_LENGTH:
-            raise ValueError(
-                f"Model {self.model} has a context window of {_ctx:,} tokens, "
-                f"which is below the minimum {MINIMUM_CONTEXT_LENGTH:,} required "
-                f"by Hermes Agent.  Choose a model with at least "
-                f"{MINIMUM_CONTEXT_LENGTH // 1000}K context, or set "
-                f"model.context_length in config.yaml to override."
-            )
-
        # Inject context engine tool schemas (e.g. lcm_grep, lcm_describe, lcm_expand)
        self._context_engine_tool_names: set = set()
        if hasattr(self, "context_compressor") and self.context_compressor and self.tools is not None:
@@ -1399,12 +1402,6 @@ class AIAgent:
            else:
                print(f"📊 Context limit: {self.context_compressor.context_length:,} tokens (auto-compression disabled)")

-        # Check immediately so CLI users see the warning at startup.
-        # Gateway status_callback is not yet wired, so any warning is stored
-        # in _compression_warning and replayed in the first run_conversation().
-        self._compression_warning = None
-        self._check_compression_model_feasibility()
-
        # Snapshot primary runtime for per-turn restoration.  When fallback
        # activates during a turn, the next turn restores these values so the
        # preferred model gets a fresh attempt each time.  Uses a single dict
@@ -1513,11 +1510,7 @@ class AIAgent:
                resolve_anthropic_token,
                _is_oauth_token,
            )
-            # Only fall back to ANTHROPIC_TOKEN when the provider is actually Anthropic.
-            # Other anthropic_messages providers (MiniMax, Alibaba, etc.) must use their own
-            # API key — falling back would send Anthropic credentials to third-party endpoints.
-            _is_native_anthropic = new_provider == "anthropic"
-            effective_key = (api_key or self.api_key or resolve_anthropic_token() or "") if _is_native_anthropic else (api_key or self.api_key or "")
+            effective_key = api_key or self.api_key or resolve_anthropic_token() or ""
            self.api_key = effective_key
            self._anthropic_api_key = effective_key
            self._anthropic_base_url = base_url or getattr(self, "_anthropic_base_url", None)
@@ -1541,7 +1534,7 @@ class AIAgent:
            )

        # ── Re-evaluate prompt caching ──
-        is_native_anthropic = api_mode == "anthropic_messages" and new_provider == "anthropic"
+        is_native_anthropic = api_mode == "anthropic_messages"
        self._use_prompt_caching = (
            ("openrouter" in (self.base_url or "").lower() and "claude" in new_model.lower())
            or is_native_anthropic
@@ -1696,104 +1689,6 @@ class AIAgent:
            except Exception:
                logger.debug("status_callback error in _emit_status", exc_info=True)

-    def _check_compression_model_feasibility(self) -> None:
-        """Warn at session start if the auxiliary compression model's context
-        window is smaller than the main model's compression threshold.
-
-        When the auxiliary model cannot fit the content that needs summarising,
-        compression will either fail outright (the LLM call errors) or produce
-        a severely truncated summary.
-
-        Called during ``__init__`` so CLI users see the warning immediately
-        (via ``_vprint``).  The gateway sets ``status_callback`` *after*
-        construction, so ``_replay_compression_warning()`` re-sends the
-        stored warning through the callback on the first
-        ``run_conversation()`` call.
-        """
-        if not self.compression_enabled:
-            return
-        try:
-            from agent.auxiliary_client import get_text_auxiliary_client
-            from agent.model_metadata import get_model_context_length
-
-            client, aux_model = get_text_auxiliary_client("compression")
-            if client is None or not aux_model:
-                msg = (
-                    "⚠ No auxiliary LLM provider configured — context "
-                    "compression will drop middle turns without a summary. "
-                    "Run `hermes setup` or set OPENROUTER_API_KEY."
-                )
-                self._compression_warning = msg
-                self._emit_status(msg)
-                logger.warning(
-                    "No auxiliary LLM provider for compression — "
-                    "summaries will be unavailable."
-                )
-                return
-
-            aux_base_url = str(getattr(client, "base_url", ""))
-            aux_api_key = str(getattr(client, "api_key", ""))
-            aux_context = get_model_context_length(
-                aux_model,
-                base_url=aux_base_url,
-                api_key=aux_api_key,
-            )
-
-            threshold = self.context_compressor.threshold_tokens
-            if aux_context < threshold:
-                # Suggest a threshold that would fit the aux model,
-                # rounded down to a clean percentage.
-                safe_pct = int((aux_context / self.context_compressor.context_length) * 100)
-                msg = (
-                    f"⚠ Compression model ({aux_model}) context "
-                    f"is {aux_context:,} tokens, but the main model's "
-                    f"compression threshold is {threshold:,} tokens. "
-                    f"Context compression will not be possible — the "
-                    f"content to summarise will exceed the auxiliary "
-                    f"model's context window.\n"
-                    f"  Fix options (config.yaml):\n"
-                    f"  1. Use a larger compression model:\n"
-                    f"       auxiliary:\n"
-                    f"         compression:\n"
-                    f"           model: <model-with-{threshold:,}+-context>\n"
-                    f"  2. Lower the compression threshold to fit "
-                    f"the current model:\n"
-                    f"       compression:\n"
-                    f"         threshold: 0.{safe_pct:02d}"
-                )
-                self._compression_warning = msg
-                self._emit_status(msg)
-                logger.warning(
-                    "Auxiliary compression model %s has %d token context, "
-                    "below the main model's compression threshold of %d "
-                    "tokens — compression summaries will fail or be "
-                    "severely truncated.",
-                    aux_model,
-                    aux_context,
-                    threshold,
-                )
-        except Exception as exc:
-            logger.debug(
-                "Compression feasibility check failed (non-fatal): %s", exc
-            )
-
-    def _replay_compression_warning(self) -> None:
-        """Re-send the compression warning through ``status_callback``.
-
-        During ``__init__`` the gateway's ``status_callback`` is not yet
-        wired, so ``_emit_status`` only reaches ``_vprint`` (CLI).  This
-        method is called once at the start of the first
-        ``run_conversation()`` — by then the gateway has set the callback,
-        so every platform (Telegram, Discord, Slack, etc.) receives the
-        warning.
-        """
-        msg = getattr(self, "_compression_warning", None)
-        if msg and self.status_callback:
-            try:
-                self.status_callback("lifecycle", msg)
-            except Exception:
-                pass
-
    def _is_direct_openai_url(self, base_url: str = None) -> bool:
        """Return True when a base URL targets OpenAI's native API."""
        url = (base_url or self._base_url_lower).lower()
@@ -1803,21 +1698,6 @@ class AIAgent:
        """Return True when the base URL targets OpenRouter."""
        return "openrouter" in self._base_url_lower

-    @staticmethod
-    def _model_requires_responses_api(model: str) -> bool:
-        """Return True for models that require the Responses API path.
-
-        GPT-5.x models are rejected on /v1/chat/completions by both
-        OpenAI and OpenRouter (error: ``unsupported_api_for_model``).
-        Detect these so the correct api_mode is set regardless of
-        which provider is serving the model.
-        """
-        m = model.lower()
-        # Strip vendor prefix (e.g. "openai/gpt-5.4" → "gpt-5.4")
-        if "/" in m:
-            m = m.rsplit("/", 1)[-1]
-        return m.startswith("gpt-5")
-
    def _max_tokens_param(self, value: int) -> dict:
        """Return the correct max tokens kwarg for the current provider.
        
@@ -2825,10 +2705,8 @@ class AIAgent:
        """
        self._interrupt_requested = True
        self._interrupt_message = message
-        # Signal all tools to abort any in-flight operations immediately.
-        # Scope the interrupt to this agent's execution thread so other
-        # agents running in the same process (gateway) are not affected.
-        _set_interrupt(True, self._execution_thread_id)
+        # Signal all tools to abort any in-flight operations immediately
+        _set_interrupt(True)
        # Propagate interrupt to any running child agents (subagent delegation)
        with self._active_children_lock:
            children_copy = list(self._active_children)
@@ -2841,10 +2719,10 @@ class AIAgent:
            print("\n⚡ Interrupt requested" + (f": '{message[:40]}...'" if message and len(message) > 40 else f": '{message}'" if message else ""))
    
    def clear_interrupt(self) -> None:
-        """Clear any pending interrupt request and the per-thread tool interrupt signal."""
+        """Clear any pending interrupt request and the global tool interrupt signal."""
        self._interrupt_requested = False
        self._interrupt_message = None
-        _set_interrupt(False, self._execution_thread_id)
+        _set_interrupt(False)

    def _touch_activity(self, desc: str) -> None:
        """Update the last-activity timestamp and description (thread-safe)."""
@@ -3182,7 +3060,7 @@ class AIAgent:
        if platform_key in PLATFORM_HINTS:
            prompt_parts.append(PLATFORM_HINTS[platform_key])

-        return "\n\n".join(p.strip() for p in prompt_parts if p.strip())
+        return "\n\n".join(prompt_parts)

    # =========================================================================
    # Pre/post-call guardrails (inspired by PR #1321 — @alireza78a)
@@ -3438,7 +3316,6 @@ class AIAgent:
    def _chat_messages_to_responses_input(self, messages: List[Dict[str, Any]]) -> List[Dict[str, Any]]:
        """Convert internal chat-style messages to Responses input items."""
        items: List[Dict[str, Any]] = []
-        seen_item_ids: set = set()

        for msg in messages:
            if not isinstance(msg, dict):
@@ -3459,12 +3336,7 @@ class AIAgent:
                    if isinstance(codex_reasoning, list):
                        for ri in codex_reasoning:
                            if isinstance(ri, dict) and ri.get("encrypted_content"):
-                                item_id = ri.get("id")
-                                if item_id and item_id in seen_item_ids:
-                                    continue
                                items.append(ri)
-                                if item_id:
-                                    seen_item_ids.add(item_id)
                                has_codex_reasoning = True

                    if content_text.strip():
@@ -3544,7 +3416,6 @@ class AIAgent:
            raise ValueError("Codex Responses input must be a list of input items.")

        normalized: List[Dict[str, Any]] = []
-        seen_ids: set = set()
        for idx, item in enumerate(raw_items):
            if not isinstance(item, dict):
                raise ValueError(f"Codex Responses input[{idx}] must be an object.")
@@ -3597,12 +3468,8 @@ class AIAgent:
            if item_type == "reasoning":
                encrypted = item.get("encrypted_content")
                if isinstance(encrypted, str) and encrypted:
-                    item_id = item.get("id")
-                    if isinstance(item_id, str) and item_id:
-                        if item_id in seen_ids:
-                            continue
-                        seen_ids.add(item_id)
                    reasoning_item = {"type": "reasoning", "encrypted_content": encrypted}
+                    item_id = item.get("id")
                    if isinstance(item_id, str) and item_id:
                        reasoning_item["id"] = item_id
                    summary = item.get("summary")
@@ -4722,49 +4589,6 @@ class AIAgent:

    # ── Unified streaming API call ─────────────────────────────────────────

-    def _reset_stream_delivery_tracking(self) -> None:
-        """Reset tracking for text delivered during the current model response."""
-        self._current_streamed_assistant_text = ""
-
-    def _record_streamed_assistant_text(self, text: str) -> None:
-        """Accumulate visible assistant text emitted through stream callbacks."""
-        if isinstance(text, str) and text:
-            self._current_streamed_assistant_text = (
-                getattr(self, "_current_streamed_assistant_text", "") + text
-            )
-
-    @staticmethod
-    def _normalize_interim_visible_text(text: str) -> str:
-        if not isinstance(text, str):
-            return ""
-        return re.sub(r"\s+", " ", text).strip()
-
-    def _interim_content_was_streamed(self, content: str) -> bool:
-        visible_content = self._normalize_interim_visible_text(
-            self._strip_think_blocks(content or "")
-        )
-        if not visible_content:
-            return False
-        streamed = self._normalize_interim_visible_text(
-            self._strip_think_blocks(getattr(self, "_current_streamed_assistant_text", "") or "")
-        )
-        return bool(streamed) and streamed == visible_content
-
-    def _emit_interim_assistant_message(self, assistant_msg: Dict[str, Any]) -> None:
-        """Surface a real mid-turn assistant commentary message to the UI layer."""
-        cb = getattr(self, "interim_assistant_callback", None)
-        if cb is None or not isinstance(assistant_msg, dict):
-            return
-        content = assistant_msg.get("content")
-        visible = self._strip_think_blocks(content or "").strip()
-        if not visible or visible == "(empty)":
-            return
-        already_streamed = self._interim_content_was_streamed(visible)
-        try:
-            cb(visible, already_streamed=already_streamed)
-        except Exception:
-            logger.debug("interim_assistant_callback error", exc_info=True)
-
    def _fire_stream_delta(self, text: str) -> None:
        """Fire all registered stream delta callbacks (display + TTS)."""
        # If a tool iteration set the break flag, prepend a single paragraph
@@ -4774,16 +4598,12 @@ class AIAgent:
        if getattr(self, "_stream_needs_break", False) and text and text.strip():
            self._stream_needs_break = False
            text = "\n\n" + text
-        callbacks = [cb for cb in (self.stream_delta_callback, self._stream_callback) if cb is not None]
-        delivered = False
-        for cb in callbacks:
-            try:
-                cb(text)
-                delivered = True
-            except Exception:
-                pass
-        if delivered:
-            self._record_streamed_assistant_text(text)
+        for cb in (self.stream_delta_callback, self._stream_callback):
+            if cb is not None:
+                try:
+                    cb(text)
+                except Exception:
+                    pass

    def _fire_reasoning_delta(self, text: str) -> None:
        """Fire reasoning callback if registered."""
@@ -4967,7 +4787,6 @@ class AIAgent:
                        if self.stream_delta_callback:
                            try:
                                self.stream_delta_callback(delta.content)
-                                self._record_streamed_assistant_text(delta.content)
                            except Exception:
                                pass

@@ -5428,7 +5247,7 @@ class AIAgent:
            except Exception:
                pass

-            # Determine api_mode from provider / base URL / model
+            # Determine api_mode from provider / base URL
            fb_api_mode = "chat_completions"
            fb_base_url = str(fb_client.base_url)
            if fb_provider == "openai-codex":
@@ -5437,10 +5256,6 @@ class AIAgent:
                fb_api_mode = "anthropic_messages"
            elif self._is_direct_openai_url(fb_base_url):
                fb_api_mode = "codex_responses"
-            elif self._model_requires_responses_api(fb_model):
-                # GPT-5.x models need Responses API on every provider
-                # (OpenRouter, Copilot, direct OpenAI, etc.)
-                fb_api_mode = "codex_responses"

            old_model = self.model
            self.model = fb_model
@@ -5482,7 +5297,7 @@ class AIAgent:
                }

            # Re-evaluate prompt caching for the new provider/model
-            is_native_anthropic = fb_api_mode == "anthropic_messages" and fb_provider == "anthropic"
+            is_native_anthropic = fb_api_mode == "anthropic_messages"
            self._use_prompt_caching = (
                ("openrouter" in fb_base_url.lower() and "claude" in fb_model.lower())
                or is_native_anthropic
@@ -5529,8 +5344,8 @@ class AIAgent:
        to the fallback provider for every subsequent turn.  Calling this at
        the top of ``run_conversation()`` makes fallback turn-scoped.

-        The gateway caches agents across messages (``_agent_cache`` in
-        ``gateway/run.py``), so this restoration IS needed there too.
+        The gateway creates a fresh agent per message so this is a no-op
+        there (``_fallback_activated`` is always False at turn start).
        """
        if not self._fallback_activated:
            return False
@@ -5818,12 +5633,11 @@ class AIAgent:
    def _anthropic_preserve_dots(self) -> bool:
        """True when using an anthropic-compatible endpoint that preserves dots in model names.
        Alibaba/DashScope keeps dots (e.g. qwen3.5-plus).
-        MiniMax keeps dots (e.g. MiniMax-M2.7).
        OpenCode Go keeps dots (e.g. minimax-m2.7)."""
-        if (getattr(self, "provider", "") or "").lower() in {"alibaba", "minimax", "minimax-cn", "opencode-go"}:
+        if (getattr(self, "provider", "") or "").lower() in {"alibaba", "opencode-go"}:
            return True
        base = (getattr(self, "base_url", "") or "").lower()
-        return "dashscope" in base or "aliyuncs" in base or "minimax" in base or "opencode.ai/zen/go" in base
+        return "dashscope" in base or "aliyuncs" in base or "opencode.ai/zen/go" in base

    def _is_qwen_portal(self) -> bool:
        """Return True when the base URL targets Qwen Portal."""
@@ -6069,16 +5883,8 @@ class AIAgent:
            api_kwargs["tools"] = self.tools

        if self.max_tokens is not None:
-            api_kwargs.update(self._max_tokens_param(self.max_tokens))
-        elif self._is_qwen_portal():
-            # Qwen Portal defaults to a very low max_tokens when omitted.
-            # Reasoning models (qwen3-coder-plus) exhaust that budget on
-            # thinking tokens alone, causing the portal to return
-            # finish_reason="stop" with truncated output — the agent sees
-            # this as an intentional stop and exits the loop.  Send 65536
-            # (the documented max output for qwen3-coder models) so the
-            # model has adequate output budget for tool calls.
-            api_kwargs.update(self._max_tokens_param(65536))
+            if not self._is_qwen_portal():
+                api_kwargs.update(self._max_tokens_param(self.max_tokens))
        elif (self._is_openrouter_url() or "nousresearch" in self._base_url_lower) and "claude" in (self.model or "").lower():
            # OpenRouter and Nous Portal translate requests to Anthropic's
            # Messages API, which requires max_tokens as a mandatory field.
@@ -6548,23 +6354,17 @@ class AIAgent:
            if messages and messages[-1].get("_flush_sentinel") == _sentinel:
                messages.pop()

-    def _compress_context(self, messages: list, system_message: str, *, approx_tokens: int = None, task_id: str = "default", focus_topic: str = None) -> tuple:
+    def _compress_context(self, messages: list, system_message: str, *, approx_tokens: int = None, task_id: str = "default") -> tuple:
        """Compress conversation context and split the session in SQLite.

-        Args:
-            focus_topic: Optional focus string for guided compression — the
-                summariser will prioritise preserving information related to
-                this topic.  Inspired by Claude Code's ``/compact <focus>``.
-
        Returns:
            (compressed_messages, new_system_prompt) tuple
        """
        _pre_msg_count = len(messages)
        logger.info(
-            "context compression started: session=%s messages=%d tokens=~%s model=%s focus=%r",
+            "context compression started: session=%s messages=%d tokens=~%s model=%s",
            self.session_id or "none", _pre_msg_count,
            f"{approx_tokens:,}" if approx_tokens else "unknown", self.model,
-            focus_topic,
        )
        # Pre-compression memory flush: let the model save memories before they're lost
        self.flush_memories(messages, min_turns=0)
@@ -6576,7 +6376,7 @@ class AIAgent:
            except Exception:
                pass

-        compressed = self.context_compressor.compress(messages, current_tokens=approx_tokens, focus_topic=focus_topic)
+        compressed = self.context_compressor.compress(messages, current_tokens=approx_tokens)

        todo_snapshot = self._todo_store.format_for_injection()
        if todo_snapshot:
@@ -6965,6 +6765,24 @@ class AIAgent:
            turn_tool_msgs = messages[-num_tools:]
            enforce_turn_budget(turn_tool_msgs, env=get_active_env(effective_task_id))

+        # ── Budget pressure injection ────────────────────────────────────
+        budget_warning = self._get_budget_warning(api_call_count)
+        if budget_warning and messages and messages[-1].get("role") == "tool":
+            last_content = messages[-1]["content"]
+            try:
+                parsed = json.loads(last_content)
+                if isinstance(parsed, dict):
+                    parsed["_budget_warning"] = budget_warning
+                    messages[-1]["content"] = json.dumps(parsed, ensure_ascii=False)
+                else:
+                    messages[-1]["content"] = last_content + f"\n\n{budget_warning}"
+            except (json.JSONDecodeError, TypeError):
+                messages[-1]["content"] = last_content + f"\n\n{budget_warning}"
+            if not self.quiet_mode:
+                remaining = self.max_iterations - api_call_count
+                tier = "⚠️  WARNING" if remaining <= self.max_iterations * 0.1 else "💡 CAUTION"
+                print(f"{self.log_prefix}{tier}: {remaining} iterations remaining")
+
    def _execute_tool_calls_sequential(self, assistant_message, messages: list, effective_task_id: str, api_call_count: int = 0) -> None:
        """Execute tool calls sequentially (original behavior). Used for single calls or interactive tools."""
        for i, tool_call in enumerate(assistant_message.tool_calls, 1):
@@ -7013,15 +6831,6 @@ class AIAgent:
            self._current_tool = function_name
            self._touch_activity(f"executing tool: {function_name}")

-            # Set activity callback for long-running tool execution (terminal
-            # commands, etc.) so the gateway's inactivity monitor doesn't kill
-            # the agent while a command is running.
-            try:
-                from tools.environments.base import set_activity_callback
-                set_activity_callback(self._touch_activity)
-            except Exception:
-                pass
-
            if self.tool_progress_callback:
                try:
                    preview = _build_tool_preview(function_name, function_args)
@@ -7311,7 +7120,50 @@ class AIAgent:
        if num_tools_seq > 0:
            enforce_turn_budget(messages[-num_tools_seq:], env=get_active_env(effective_task_id))

+        # ── Budget pressure injection ─────────────────────────────────
+        # After all tool calls in this turn are processed, check if we're
+        # approaching max_iterations. If so, inject a warning into the LAST
+        # tool result's JSON so the LLM sees it naturally when reading results.
+        budget_warning = self._get_budget_warning(api_call_count)
+        if budget_warning and messages and messages[-1].get("role") == "tool":
+            last_content = messages[-1]["content"]
+            try:
+                parsed = json.loads(last_content)
+                if isinstance(parsed, dict):
+                    parsed["_budget_warning"] = budget_warning
+                    messages[-1]["content"] = json.dumps(parsed, ensure_ascii=False)
+                else:
+                    messages[-1]["content"] = last_content + f"\n\n{budget_warning}"
+            except (json.JSONDecodeError, TypeError):
+                messages[-1]["content"] = last_content + f"\n\n{budget_warning}"
+            if not self.quiet_mode:
+                remaining = self.max_iterations - api_call_count
+                tier = "⚠️  WARNING" if remaining <= self.max_iterations * 0.1 else "💡 CAUTION"
+                print(f"{self.log_prefix}{tier}: {remaining} iterations remaining")

+    def _get_budget_warning(self, api_call_count: int) -> Optional[str]:
+        """Return a budget pressure string, or None if not yet needed.
+
+        Two-tier system:
+          - Caution (70%): nudge to consolidate work
+          - Warning (90%): urgent, must respond now
+        """
+        if not self._budget_pressure_enabled or self.max_iterations <= 0:
+            return None
+        progress = api_call_count / self.max_iterations
+        remaining = self.max_iterations - api_call_count
+        if progress >= self._budget_warning_threshold:
+            return (
+                f"[BUDGET WARNING: Iteration {api_call_count}/{self.max_iterations}. "
+                f"Only {remaining} iteration(s) left. "
+                "Provide your final response NOW. No more tool calls unless absolutely critical.]"
+            )
+        if progress >= self._budget_caution_threshold:
+            return (
+                f"[BUDGET: Iteration {api_call_count}/{self.max_iterations}. "
+                f"{remaining} iterations left. Start consolidating your work.]"
+            )
+        return None

    def _emit_context_pressure(self, compaction_progress: float, compressor) -> None:
        """Notify the user that context is approaching the compaction threshold.
@@ -7535,11 +7387,6 @@ class AIAgent:
        # Installed once, transparent when streams are healthy, prevents crash on write.
        _install_safe_stdio()

-        # Tag all log records on this thread with the session ID so
-        # ``hermes logs --session <id>`` can filter a single conversation.
-        from hermes_logging import set_session_context
-        set_session_context(self.session_id)
-
        # If the previous turn activated fallback, restore the primary
        # runtime so this turn gets a fresh attempt with the preferred model.
        # No-op when _fallback_activated is False (gateway, first turn, etc.).
@@ -7585,12 +7432,6 @@ class AIAgent:
                    )
            except Exception:
                pass
-        # Replay compression warning through status_callback for gateway
-        # platforms (the callback was not wired during __init__).
-        if self._compression_warning:
-            self._replay_compression_warning()
-            self._compression_warning = None  # send once
-
        # NOTE: _turns_since_memory and _iters_since_skill are NOT reset here.
        # They are initialized in __init__ and must persist across run_conversation
        # calls so that nudge logic accumulates correctly in CLI mode.
@@ -7609,6 +7450,14 @@ class AIAgent:
        # Initialize conversation (copy to avoid mutating the caller's list)
        messages = list(conversation_history) if conversation_history else []

+        # Strip budget pressure warnings from previous turns.  These are
+        # turn-scoped signals injected by _get_budget_warning() into tool
+        # result content.  If left in the replayed history, models (especially
+        # GPT-family) interpret them as still-active instructions and avoid
+        # making tool calls in ALL subsequent turns.
+        if messages:
+            _strip_budget_warnings_from_history(messages)
+        
        # Hydrate todo store from conversation history (gateway creates a fresh
        # AIAgent per message, so the in-memory store is empty -- we need to
        # recover the todo state from the most recent todo tool response in history)
@@ -7804,11 +7653,6 @@ class AIAgent:
        compression_attempts = 0
        _turn_exit_reason = "unknown"  # Diagnostic: why the loop ended
        
-        # Record the execution thread so interrupt()/clear_interrupt() can
-        # scope the tool-level interrupt signal to THIS agent's thread only.
-        # Must be set before clear_interrupt() which uses it.
-        self._execution_thread_id = threading.current_thread().ident
-
        # Clear any stale interrupt state at start
        self.clear_interrupt()

@@ -7825,7 +7669,7 @@ class AIAgent:
            except Exception:
                pass

-        while (api_call_count < self.max_iterations and self.iteration_budget.remaining > 0) or self._budget_grace_call:
+        while api_call_count < self.max_iterations and self.iteration_budget.remaining > 0:
            # Reset per-turn checkpoint dedup so each iteration can take one snapshot
            self._checkpoint_mgr.new_turn()

@@ -7840,13 +7684,7 @@ class AIAgent:
            api_call_count += 1
            self._api_call_count = api_call_count
            self._touch_activity(f"starting API call #{api_call_count}")
-
-            # Grace call: the budget is exhausted but we gave the model one
-            # more chance.  Consume the grace flag so the loop exits after
-            # this iteration regardless of outcome.
-            if self._budget_grace_call:
-                self._budget_grace_call = False
-            elif not self.iteration_budget.consume():
+            if not self.iteration_budget.consume():
                _turn_exit_reason = "budget_exhausted"
                if not self.quiet_mode:
                    self._safe_print(f"\n⚠️  Iteration budget exhausted ({self.iteration_budget.used}/{self.iteration_budget.max_total} iterations used)")
@@ -7973,39 +7811,9 @@ class AIAgent:
            # manual message manipulation are always caught.
            api_messages = self._sanitize_api_messages(api_messages)

-            # Normalize message whitespace and tool-call JSON for consistent
-            # prefix matching.  Ensures bit-perfect prefixes across turns,
-            # which enables KV cache reuse on local inference servers
-            # (llama.cpp, vLLM, Ollama) and improves cache hit rates for
-            # cloud providers.  Operates on api_messages (the API copy) so
-            # the original conversation history in `messages` is untouched.
-            for am in api_messages:
-                if isinstance(am.get("content"), str):
-                    am["content"] = am["content"].strip()
-            for am in api_messages:
-                tcs = am.get("tool_calls")
-                if not tcs:
-                    continue
-                new_tcs = []
-                for tc in tcs:
-                    if isinstance(tc, dict) and "function" in tc:
-                        try:
-                            args_obj = json.loads(tc["function"]["arguments"])
-                            tc = {**tc, "function": {
-                                **tc["function"],
-                                "arguments": json.dumps(
-                                    args_obj, separators=(",", ":"),
-                                    sort_keys=True,
-                                ),
-                            }}
-                        except Exception:
-                            pass
-                    new_tcs.append(tc)
-                am["tool_calls"] = new_tcs
-
            # Calculate approximate request size for logging
            total_chars = sum(len(str(msg)) for msg in api_messages)
-            approx_tokens = estimate_messages_tokens_rough(api_messages)
+            approx_tokens = total_chars // 4  # Rough estimate: 4 chars per token
            
            # Thinking spinner for quiet mode (animated during API call)
            thinking_spinner = None
@@ -8054,7 +7862,6 @@ class AIAgent:

            while retry_count < max_retries:
                try:
-                    self._reset_stream_delivery_tracking()
                    api_kwargs = self._build_api_kwargs(api_messages)
                    if self.api_mode == "codex_responses":
                        api_kwargs = self._preflight_codex_api_kwargs(api_kwargs, allow_stream=False)
@@ -8213,8 +8020,6 @@ class AIAgent:
                            self._emit_status("⚠️ Empty/malformed response — switching to fallback...")
                        if self._try_activate_fallback():
                            retry_count = 0
-                            compression_attempts = 0
-                            primary_recovery_attempted = False
                            continue

                        # Check for error field in response (some providers include this)
@@ -8250,8 +8055,6 @@ class AIAgent:
                            self._emit_status(f"⚠️ Max retries ({max_retries}) for invalid responses — trying fallback...")
                            if self._try_activate_fallback():
                                retry_count = 0
-                                compression_attempts = 0
-                                primary_recovery_attempted = False
                                continue
                            self._emit_status(f"❌ Max retries ({max_retries}) exceeded for invalid responses. Giving up.")
                            logging.error(f"{self.log_prefix}Invalid API response after {max_retries} retries.")
@@ -8328,24 +8131,8 @@ class AIAgent:
                                    _text_parts.append(getattr(_blk, "text", ""))
                            _trunc_content = "\n".join(_text_parts) if _text_parts else None

-                        # A response is "thinking exhausted" only when the model
-                        # actually produced reasoning blocks but no visible text after
-                        # them.  Models that do not use <think> tags (e.g. GLM-4.7 on
-                        # NVIDIA Build, minimax) may return content=None or an empty
-                        # string for unrelated reasons — treat those as normal
-                        # truncations that deserve continuation retries, not as
-                        # thinking-budget exhaustion.
-                        _has_think_tags = bool(
-                            _trunc_content and re.search(
-                                r'<(?:think|thinking|reasoning|REASONING_SCRATCHPAD)[^>]*>',
-                                _trunc_content,
-                                re.IGNORECASE,
-                            )
-                        )
                        _thinking_exhausted = (
-                            not _trunc_has_tool_calls
-                            and _has_think_tags
-                            and (
+                            not _trunc_has_tool_calls and (
                                (_trunc_content is not None and not self._has_content_after_think_block(_trunc_content))
                                or _trunc_content is None
                            )
@@ -8906,8 +8693,6 @@ class AIAgent:
                            self._emit_status("⚠️ Rate limited — switching to fallback provider...")
                            if self._try_activate_fallback():
                                retry_count = 0
-                                compression_attempts = 0
-                                primary_recovery_attempted = False
                                continue

                    is_payload_too_large = (
@@ -9121,8 +8906,6 @@ class AIAgent:
                        self._emit_status(f"⚠️ Non-retryable error (HTTP {status_code}) — trying fallback...")
                        if self._try_activate_fallback():
                            retry_count = 0
-                            compression_attempts = 0
-                            primary_recovery_attempted = False
                            continue
                        if api_kwargs is not None:
                            self._dump_api_request_debug(
@@ -9188,8 +8971,6 @@ class AIAgent:
                        self._emit_status(f"⚠️ Max retries ({max_retries}) exhausted — trying fallback...")
                        if self._try_activate_fallback():
                            retry_count = 0
-                            compression_attempts = 0
-                            primary_recovery_attempted = False
                            continue
                        _final_summary = self._summarize_api_error(api_error)
                        if is_rate_limited:
@@ -9413,6 +9194,8 @@ class AIAgent:
                # Check for incomplete <REASONING_SCRATCHPAD> (opened but never closed)
                # This means the model ran out of output tokens mid-reasoning — retry up to 2 times
                if has_incomplete_scratchpad(assistant_message.content or ""):
+                    if not hasattr(self, '_incomplete_scratchpad_retries'):
+                        self._incomplete_scratchpad_retries = 0
                    self._incomplete_scratchpad_retries += 1
                    
                    self._vprint(f"{self.log_prefix}⚠️  Incomplete <REASONING_SCRATCHPAD> detected (opened but never closed)")
@@ -9440,9 +9223,12 @@ class AIAgent:
                        }
                
                # Reset incomplete scratchpad counter on clean response
-                self._incomplete_scratchpad_retries = 0
+                if hasattr(self, '_incomplete_scratchpad_retries'):
+                    self._incomplete_scratchpad_retries = 0

                if self.api_mode == "codex_responses" and finish_reason == "incomplete":
+                    if not hasattr(self, "_codex_incomplete_retries"):
+                        self._codex_incomplete_retries = 0
                    self._codex_incomplete_retries += 1

                    interim_msg = self._build_assistant_message(assistant_message, finish_reason)
@@ -9469,7 +9255,6 @@ class AIAgent:
                        )
                        if not duplicate_interim:
                            messages.append(interim_msg)
-                            self._emit_interim_assistant_message(interim_msg)

                    if self._codex_incomplete_retries < 3:
                        if not self.quiet_mode:
@@ -9514,6 +9299,8 @@ class AIAgent:
                    ]
                    if invalid_tool_calls:
                        # Track retries for invalid tool calls
+                        if not hasattr(self, '_invalid_tool_retries'):
+                            self._invalid_tool_retries = 0
                        self._invalid_tool_retries += 1

                        # Return helpful error to model — model can self-correct next turn
@@ -9549,7 +9336,8 @@ class AIAgent:
                            })
                        continue
                    # Reset retry counter on successful tool call validation
-                    self._invalid_tool_retries = 0
+                    if hasattr(self, '_invalid_tool_retries'):
+                        self._invalid_tool_retries = 0
                    
                    # Validate tool call arguments are valid JSON
                    # Handle empty strings as empty objects (common model quirk)
@@ -9572,41 +9360,12 @@ class AIAgent:
                            invalid_json_args.append((tc.function.name, str(e)))
                    
                    if invalid_json_args:
-                        # Check if the invalid JSON is due to truncation rather
-                        # than a model formatting mistake.  Routers sometimes
-                        # rewrite finish_reason from "length" to "tool_calls",
-                        # hiding the truncation from the length handler above.
-                        # Detect truncation: args that don't end with } or ]
-                        # (after stripping whitespace) are cut off mid-stream.
-                        _truncated = any(
-                            not (tc.function.arguments or "").rstrip().endswith(("}", "]"))
-                            for tc in assistant_message.tool_calls
-                            if tc.function.name in {n for n, _ in invalid_json_args}
-                        )
-                        if _truncated:
-                            self._vprint(
-                                f"{self.log_prefix}⚠️  Truncated tool call arguments detected "
-                                f"(finish_reason={finish_reason!r}) — refusing to execute.",
-                                force=True,
-                            )
-                            self._invalid_json_retries = 0
-                            self._cleanup_task_resources(effective_task_id)
-                            self._persist_session(messages, conversation_history)
-                            return {
-                                "final_response": None,
-                                "messages": messages,
-                                "api_calls": api_call_count,
-                                "completed": False,
-                                "partial": True,
-                                "error": "Response truncated due to output length limit",
-                            }
-
                        # Track retries for invalid JSON arguments
                        self._invalid_json_retries += 1
-
+                        
                        tool_name, error_msg = invalid_json_args[0]
                        self._vprint(f"{self.log_prefix}⚠️  Invalid JSON in tool call arguments for '{tool_name}': {error_msg}")
-
+                        
                        if self._invalid_json_retries < 3:
                            self._vprint(f"{self.log_prefix}🔄 Retrying API call ({self._invalid_json_retries}/3)...")
                            # Don't add anything to messages, just retry the API call
@@ -9689,7 +9448,6 @@ class AIAgent:
                        messages.pop()

                    messages.append(assistant_msg)
-                    self._emit_interim_assistant_message(assistant_msg)

                    # Close any open streaming display (response box, reasoning
                    # box) before tool execution begins.  Intermediate turns may
@@ -9952,7 +9710,8 @@ class AIAgent:
                        break
                    
                    # Reset retry counter/signature on successful content
-                    self._empty_content_retries = 0
+                    if hasattr(self, '_empty_content_retries'):
+                        self._empty_content_retries = 0
                    self._thinking_prefill_retries = 0

                    if (
@@ -9968,7 +9727,6 @@ class AIAgent:
                        codex_ack_continuations += 1
                        interim_msg = self._build_assistant_message(assistant_message, "incomplete")
                        messages.append(interim_msg)
-                        self._emit_interim_assistant_message(interim_msg)

                        continue_msg = {
                            "role": "user",
@@ -10019,7 +9777,8 @@ class AIAgent:
                except (OSError, ValueError):
                    logger.error(error_msg)
                
-                logger.debug("Outer loop error in API call #%d", api_call_count, exc_info=True)
+                if self.verbose_logging:
+                    logging.exception("Detailed error information:")
                
                # If an assistant message with tool_calls was already appended,
                # the API expects a role="tool" result for every tool_call_id.
@@ -10065,31 +9824,7 @@ class AIAgent:
        if final_response is None and (
            api_call_count >= self.max_iterations
            or self.iteration_budget.remaining <= 0
-        ) and not self._budget_exhausted_injected:
-            # Budget exhausted but we haven't tried asking the model to
-            # summarise yet.  Inject a user message and give it one grace
-            # API call to produce a text response.
-            self._budget_exhausted_injected = True
-            self._budget_grace_call = True
-            _grace_msg = (
-                "Your tool budget ran out. Please give me the information "
-                "or actions you've completed so far."
-            )
-            messages.append({"role": "user", "content": _grace_msg})
-            self._emit_status(
-                f"⚠️ Iteration budget exhausted ({api_call_count}/{self.max_iterations}) "
-                "— asking model to summarise"
-            )
-            if not self.quiet_mode:
-                self._safe_print(
-                    f"\n⚠️  Iteration budget exhausted ({api_call_count}/{self.max_iterations}) "
-                    "— requesting summary..."
-                )
-
-        if final_response is None and (
-            api_call_count >= self.max_iterations
-            or self.iteration_budget.remaining <= 0
-        ) and not self._budget_grace_call:
+        ):
            _turn_exit_reason = f"max_iterations_reached({api_call_count}/{self.max_iterations})"
            if self.iteration_budget.remaining <= 0 and not self.quiet_mode:
                print(f"\n⚠️  Iteration budget exhausted ({self.iteration_budget.used}/{self.iteration_budget.max_total} iterations used)")
@@ -249,12 +249,8 @@ def check_config(groq_key, eleven_key):

            if stt_provider == "groq" and not groq_key:
                warn("STT config says groq but GROQ_API_KEY is missing")
-            if stt_provider == "mistral" and not os.getenv("MISTRAL_API_KEY"):
-                warn("STT config says mistral but MISTRAL_API_KEY is missing")
            if tts_provider == "elevenlabs" and not eleven_key:
                warn("TTS config says elevenlabs but ELEVENLABS_API_KEY is missing")
-            if tts_provider == "mistral" and not os.getenv("MISTRAL_API_KEY"):
-                warn("TTS config says mistral but MISTRAL_API_KEY is missing")
        except Exception as e:
            warn("config.yaml", f"parse error: {e}")
    else:
@@ -8,7 +8,7 @@
      "name": "hermes-whatsapp-bridge",
      "version": "1.0.0",
      "dependencies": {
-        "@whiskeysockets/baileys": "WhiskeySockets/Baileys#fix/abprops-abt-fetch",
+        "@whiskeysockets/baileys": "7.0.0-rc.9",
        "express": "^4.21.0",
        "pino": "^9.0.0",
        "qrcode-terminal": "^0.12.0"
@@ -730,22 +730,21 @@
      }
    },
    "node_modules/@whiskeysockets/baileys": {
-      "name": "baileys",
      "version": "7.0.0-rc.9",
-      "resolved": "git+ssh://git@github.com/WhiskeySockets/Baileys.git#01047debd81beb20da7b7779b08edcb06aa03770",
+      "resolved": "https://registry.npmjs.org/@whiskeysockets/baileys/-/baileys-7.0.0-rc.9.tgz",
+      "integrity": "sha512-YFm5gKXfDP9byCXCW3OPHKXLzrAKzolzgVUlRosHHgwbnf2YOO3XknkMm6J7+F0ns8OA0uuSBhgkRHTDtqkacw==",
      "hasInstallScript": true,
      "license": "MIT",
      "dependencies": {
        "@cacheable/node-cache": "^1.4.0",
        "@hapi/boom": "^9.1.3",
        "async-mutex": "^0.5.0",
-        "libsignal": "git+https://github.com/whiskeysockets/libsignal-node",
+        "libsignal": "git+https://github.com/whiskeysockets/libsignal-node.git",
        "lru-cache": "^11.1.0",
        "music-metadata": "^11.7.0",
        "p-queue": "^9.0.0",
        "pino": "^9.6",
        "protobufjs": "^7.2.4",
-        "whatsapp-rust-bridge": "0.5.2",
        "ws": "^8.13.0"
      },
      "engines": {
@@ -2126,12 +2125,6 @@
        "node": ">= 0.8"
      }
    },
-    "node_modules/whatsapp-rust-bridge": {
-      "version": "0.5.2",
-      "resolved": "https://registry.npmjs.org/whatsapp-rust-bridge/-/whatsapp-rust-bridge-0.5.2.tgz",
-      "integrity": "sha512-6KBRNvxg6WMIwZ/euA8qVzj16qxMBzLllfmaJIP1JGAAfSvwn6nr8JDOMXeqpXPEOl71UfOG+79JwKEoT2b1Fw==",
-      "license": "MIT"
-    },
    "node_modules/win-guid": {
      "version": "0.2.1",
      "resolved": "https://registry.npmjs.org/win-guid/-/win-guid-0.2.1.tgz",
@@ -8,7 +8,7 @@
    "start": "node bridge.js"
  },
  "dependencies": {
-    "@whiskeysockets/baileys": "WhiskeySockets/Baileys#fix/abprops-abt-fetch",
+    "@whiskeysockets/baileys": "7.0.0-rc.9",
    "express": "^4.21.0",
    "qrcode-terminal": "^0.12.0",
    "pino": "^9.0.0"
@@ -1,29 +1,51 @@
 ---
 name: github-code-review
-description: Review code changes by analyzing git diffs, leaving inline comments on PRs, and performing thorough pre-push review. Uses GitHub MCP tools (mcp_github_*) as the primary interface, with git CLI for local diff operations.
-version: 2.0.0
+description: Review code changes by analyzing git diffs, leaving inline comments on PRs, and performing thorough pre-push review. Works with gh CLI or falls back to git + GitHub REST API via curl.
+version: 1.1.0
 author: Hermes Agent
 license: MIT
 metadata:
  hermes:
-    tags: [GitHub, Code-Review, Pull-Requests, Git, Quality, MCP]
+    tags: [GitHub, Code-Review, Pull-Requests, Git, Quality]
    related_skills: [github-auth, github-pr-workflow]
 ---

 # GitHub Code Review

-Perform code reviews on local changes before pushing, or review open PRs on GitHub. This skill uses **GitHub MCP tools** (`mcp_github_*`) as the primary interface for all GitHub API interactions, with plain `git` for local diff operations.
+Perform code reviews on local changes before pushing, or review open PRs on GitHub. Most of this skill uses plain `git` — the `gh`/`curl` split only matters for PR-level interactions.

 ## Prerequisites

- GitHub MCP server configured (provides `mcp_github_*` tools)
- Inside a git repository (for local diff operations)
+- Authenticated with GitHub (see `github-auth` skill)
+- Inside a git repository
+
+### Setup (for PR interactions)
+
+```bash
+if command -v gh &>/dev/null && gh auth status &>/dev/null; then
+  AUTH="gh"
+else
+  AUTH="git"
+  if [ -z "$GITHUB_TOKEN" ]; then
+    if [ -f ~/.hermes/.env ] && grep -q "^GITHUB_TOKEN=" ~/.hermes/.env; then
+      GITHUB_TOKEN=$(grep "^GITHUB_TOKEN=" ~/.hermes/.env | head -1 | cut -d= -f2 | tr -d '\n\r')
+    elif grep -q "github.com" ~/.git-credentials 2>/dev/null; then
+      GITHUB_TOKEN=$(grep "github.com" ~/.git-credentials 2>/dev/null | head -1 | sed 's|https://[^:]*:\([^@]*\)@.*|\1|')
+    fi
+  fi
+fi
+
+REMOTE_URL=$(git remote get-url origin)
+OWNER_REPO=$(echo "$REMOTE_URL" | sed -E 's|.*github\.com[:/]||; s|\.git$||')
+OWNER=$(echo "$OWNER_REPO" | cut -d/ -f1)
+REPO=$(echo "$OWNER_REPO" | cut -d/ -f2)
+```

 ---

 ## 1. Reviewing Local Changes (Pre-Push)

-Local diffs use plain `git` — no API needed.
+This is pure `git` — works everywhere, no API needed.

 ### Get the Diff

@@ -100,206 +122,158 @@ When reviewing local changes, present findings in this structure:

 ---

-## 2. Reviewing a Pull Request on GitHub (MCP Tools)
+## 2. Reviewing a Pull Request on GitHub

-### Step 1: Gather PR Context
+### View PR Details

-Use MCP tools to get PR metadata, description, and changed files:
-
-```
-# Get PR details (title, author, description, branch, status)
-mcp_github_pull_request_read(method="get", owner=OWNER, repo=REPO, pullNumber=PR_NUMBER)
-
-# Get the diff
-mcp_github_pull_request_read(method="get_diff", owner=OWNER, repo=REPO, pullNumber=PR_NUMBER)
-
-# Get list of changed files with additions/deletions
-mcp_github_pull_request_read(method="get_files", owner=OWNER, repo=REPO, pullNumber=PR_NUMBER)
-
-# Get CI/CD status
-mcp_github_pull_request_read(method="get_status", owner=OWNER, repo=REPO, pullNumber=PR_NUMBER)
-
-# Get check runs (individual CI jobs)
-mcp_github_pull_request_read(method="get_check_runs", owner=OWNER, repo=REPO, pullNumber=PR_NUMBER)
-```
-
-### Step 2: Read File Contents for Context
-
-For each changed file, read the full file to understand the surrounding context:
-
-```
-# Read specific files from the PR branch
-mcp_github_get_file_contents(owner=OWNER, repo=REPO, path="src/auth/login.py", ref="refs/pull/PR_NUMBER/head")
-```
-
-### Step 3: Check Out Locally (Optional — for running tests)
-
-If you need to run tests or linters locally:
+**With gh:**

 ```bash
-git fetch origin pull/PR_NUMBER/head:pr-PR_NUMBER
-git checkout pr-PR_NUMBER
-
-# Run tests
-python -m pytest 2>&1 | tail -20
-
-# Run linter
-ruff check . 2>&1 | head -30
+gh pr view 123
+gh pr diff 123
+gh pr diff 123 --name-only
 ```

-### Step 4: Get Existing Review Comments
-
-Check what's already been discussed:
-
-```
-# Get review threads (grouped comments on code locations)
-mcp_github_pull_request_read(method="get_review_comments", owner=OWNER, repo=REPO, pullNumber=PR_NUMBER)
-
-# Get general PR comments
-mcp_github_pull_request_read(method="get_comments", owner=OWNER, repo=REPO, pullNumber=PR_NUMBER)
-
-# Get formal reviews (approvals, change requests)
-mcp_github_pull_request_read(method="get_reviews", owner=OWNER, repo=REPO, pullNumber=PR_NUMBER)
-```
-
-### Step 5: Apply the Review Checklist (Section 3)
-
-Go through each category systematically.
-
-### Step 6: Submit a Formal Review with Inline Comments
-
-Use the MCP review tools to submit findings:
-
-**Create a pending review, add inline comments, then submit:**
-
-```
-# Step A: Create a pending review (omit "event" to keep it pending)
-mcp_github_pull_request_review_write(
-    method="create",
-    owner=OWNER,
-    repo=REPO,
-    pullNumber=PR_NUMBER
-)
-
-# Step B: Add inline comments to the pending review
-mcp_github_add_comment_to_pending_review(
-    owner=OWNER,
-    repo=REPO,
-    pullNumber=PR_NUMBER,
-    path="src/auth.py",
-    line=45,
-    body="🔴 **Critical:** User input passed directly to SQL query — use parameterized queries.",
-    subjectType="LINE",
-    side="RIGHT"
-)
-
-mcp_github_add_comment_to_pending_review(
-    owner=OWNER,
-    repo=REPO,
-    pullNumber=PR_NUMBER,
-    path="src/models/user.py",
-    line=23,
-    body="⚠️ **Warning:** Password stored without hashing. Use bcrypt or argon2.",
-    subjectType="LINE",
-    side="RIGHT"
-)
-
-# Step C: Submit the pending review
-mcp_github_pull_request_review_write(
-    method="submit_pending",
-    owner=OWNER,
-    repo=REPO,
-    pullNumber=PR_NUMBER,
-    event="REQUEST_CHANGES",  # or "APPROVE" or "COMMENT"
-    body="## Hermes Agent Review\n\nFound 2 issues. See inline comments."
-)
-```
-
-**Or submit a review directly (no pending step):**
-
-```
-# Approve
-mcp_github_pull_request_review_write(
-    method="create",
-    owner=OWNER,
-    repo=REPO,
-    pullNumber=PR_NUMBER,
-    event="APPROVE",
-    body="LGTM! Code looks clean — good test coverage, no security concerns."
-)
-
-# Request changes
-mcp_github_pull_request_review_write(
-    method="create",
-    owner=OWNER,
-    repo=REPO,
-    pullNumber=PR_NUMBER,
-    event="REQUEST_CHANGES",
-    body="Found a few issues — see inline comments."
-)
-```
-
-### Step 7: Post a Summary Comment
-
-Leave a top-level summary so the PR author gets the full picture:
-
-```
-mcp_github_add_issue_comment(
-    owner=OWNER,
-    repo=REPO,
-    issue_number=PR_NUMBER,
-    body="""## Code Review Summary
-
-**Verdict: Changes Requested** (2 issues, 1 suggestion)
-
-### 🔴 Critical
- **src/auth.py:45** — SQL injection vulnerability
-
-### ⚠️ Warnings
- **src/models.py:23** — Plaintext password storage
-
-### 💡 Suggestions
- **src/utils.py:8** — Duplicated logic, consider consolidating
-
-### ✅ Looks Good
- Clean API design
- Good error handling in the middleware layer
-
---
-*Reviewed by Hermes Agent*"""
-)
-```
-
-### Step 8: Reply to Existing Comments
-
-If the PR author responds to your review:
-
-```
-# Reply to a specific review comment
-mcp_github_add_reply_to_pull_request_comment(
-    owner=OWNER,
-    repo=REPO,
-    pullNumber=PR_NUMBER,
-    commentId=COMMENT_ID,
-    body="Good point! That approach works too."
-)
-```
-
-### Step 9: Request Copilot Review (Optional)
-
-For automated AI feedback before your review:
-
-```
-mcp_github_request_copilot_review(owner=OWNER, repo=REPO, pullNumber=PR_NUMBER)
-```
-
-### Step 10: Clean Up (if checked out locally)
+**With git + curl:**

 ```bash
-git checkout main
-git branch -D pr-PR_NUMBER
+PR_NUMBER=123
+
+# Get PR details
+curl -s \
+  -H "Authorization: token $GITHUB_TOKEN" \
+  https://api.github.com/repos/$OWNER/$REPO/pulls/$PR_NUMBER \
+  | python3 -c "
+import sys, json
+pr = json.load(sys.stdin)
+print(f\"Title: {pr['title']}\")
+print(f\"Author: {pr['user']['login']}\")
+print(f\"Branch: {pr['head']['ref']} -> {pr['base']['ref']}\")
+print(f\"State: {pr['state']}\")
+print(f\"Body:\n{pr['body']}\")"
+
+# List changed files
+curl -s \
+  -H "Authorization: token $GITHUB_TOKEN" \
+  https://api.github.com/repos/$OWNER/$REPO/pulls/$PR_NUMBER/files \
+  | python3 -c "
+import sys, json
+for f in json.load(sys.stdin):
+    print(f\"{f['status']:10} +{f['additions']:-4} -{f['deletions']:-4}  {f['filename']}\")"
 ```

+### Check Out PR Locally for Full Review
+
+This works with plain `git` — no `gh` needed:
+
+```bash
+# Fetch the PR branch and check it out
+git fetch origin pull/123/head:pr-123
+git checkout pr-123
+
+# Now you can use read_file, search_files, run tests, etc.
+
+# View diff against the base branch
+git diff main...pr-123
+```
+
+**With gh (shortcut):**
+
+```bash
+gh pr checkout 123
+```
+
+### Leave Comments on a PR
+
+**General PR comment — with gh:**
+
+```bash
+gh pr comment 123 --body "Overall looks good, a few suggestions below."
+```
+
+**General PR comment — with curl:**
+
+```bash
+curl -s -X POST \
+  -H "Authorization: token $GITHUB_TOKEN" \
+  https://api.github.com/repos/$OWNER/$REPO/issues/$PR_NUMBER/comments \
+  -d '{"body": "Overall looks good, a few suggestions below."}'
+```
+
+### Leave Inline Review Comments
+
+**Single inline comment — with gh (via API):**
+
+```bash
+HEAD_SHA=$(gh pr view 123 --json headRefOid --jq '.headRefOid')
+
+gh api repos/$OWNER/$REPO/pulls/123/comments \
+  --method POST \
+  -f body="This could be simplified with a list comprehension." \
+  -f path="src/auth/login.py" \
+  -f commit_id="$HEAD_SHA" \
+  -f line=45 \
+  -f side="RIGHT"
+```
+
+**Single inline comment — with curl:**
+
+```bash
+# Get the head commit SHA
+HEAD_SHA=$(curl -s \
+  -H "Authorization: token $GITHUB_TOKEN" \
+  https://api.github.com/repos/$OWNER/$REPO/pulls/$PR_NUMBER \
+  | python3 -c "import sys,json; print(json.load(sys.stdin)['head']['sha'])")
+
+curl -s -X POST \
+  -H "Authorization: token $GITHUB_TOKEN" \
+  https://api.github.com/repos/$OWNER/$REPO/pulls/$PR_NUMBER/comments \
+  -d "{
+    \"body\": \"This could be simplified with a list comprehension.\",
+    \"path\": \"src/auth/login.py\",
+    \"commit_id\": \"$HEAD_SHA\",
+    \"line\": 45,
+    \"side\": \"RIGHT\"
+  }"
+```
+
+### Submit a Formal Review (Approve / Request Changes)
+
+**With gh:**
+
+```bash
+gh pr review 123 --approve --body "LGTM!"
+gh pr review 123 --request-changes --body "See inline comments."
+gh pr review 123 --comment --body "Some suggestions, nothing blocking."
+```
+
+**With curl — multi-comment review submitted atomically:**
+
+```bash
+HEAD_SHA=$(curl -s \
+  -H "Authorization: token $GITHUB_TOKEN" \
+  https://api.github.com/repos/$OWNER/$REPO/pulls/$PR_NUMBER \
+  | python3 -c "import sys,json; print(json.load(sys.stdin)['head']['sha'])")
+
+curl -s -X POST \
+  -H "Authorization: token $GITHUB_TOKEN" \
+  https://api.github.com/repos/$OWNER/$REPO/pulls/$PR_NUMBER/reviews \
+  -d "{
+    \"commit_id\": \"$HEAD_SHA\",
+    \"event\": \"COMMENT\",
+    \"body\": \"Code review from Hermes Agent\",
+    \"comments\": [
+      {\"path\": \"src/auth.py\", \"line\": 45, \"body\": \"Use parameterized queries to prevent SQL injection.\"},
+      {\"path\": \"src/models/user.py\", \"line\": 23, \"body\": \"Hash passwords with bcrypt before storing.\"},
+      {\"path\": \"tests/test_auth.py\", \"line\": 1, \"body\": \"Add test for expired token edge case.\"}
+    ]
+  }"
+```
+
+Event values: `"APPROVE"`, `"REQUEST_CHANGES"`, `"COMMENT"`
+
+The `line` field refers to the line number in the *new* version of the file. For deleted lines, use `"side": "LEFT"`.
+
 ---

 ## 3. Review Checklist
@@ -316,7 +290,6 @@ When performing a code review (local or PR), systematically check:
 - Input validation on user-facing inputs
 - No SQL injection, XSS, or path traversal
 - Auth/authz checks where needed
- Use `mcp_github_run_secret_scanning` on changed files for automated secret detection

 ### Code Quality
 - Clear naming (variables, functions, classes)
@@ -354,30 +327,151 @@ When the user asks you to "review the code" or "check before pushing":

 ---

-## 5. PR Review Workflow (End-to-End with MCP Tools)
+## 5. PR Review Workflow (End-to-End)

-When the user asks you to "review PR #N", "look at this PR", or gives you a PR URL:
+When the user asks you to "review PR #N", "look at this PR", or gives you a PR URL, follow this recipe:

-### Quick Reference
+### Step 1: Set up environment

-| Task | MCP Tool |
-|------|----------|
-| Get PR details | `mcp_github_pull_request_read(method="get")` |
-| Get PR diff | `mcp_github_pull_request_read(method="get_diff")` |
-| Get changed files | `mcp_github_pull_request_read(method="get_files")` |
-| Get CI status | `mcp_github_pull_request_read(method="get_status")` |
-| Get check runs | `mcp_github_pull_request_read(method="get_check_runs")` |
-| Read file contents | `mcp_github_get_file_contents(ref="refs/pull/N/head")` |
-| Get review threads | `mcp_github_pull_request_read(method="get_review_comments")` |
-| Get PR comments | `mcp_github_pull_request_read(method="get_comments")` |
-| Get reviews | `mcp_github_pull_request_read(method="get_reviews")` |
-| Create pending review | `mcp_github_pull_request_review_write(method="create")` |
-| Add inline comment | `mcp_github_add_comment_to_pending_review()` |
-| Submit review | `mcp_github_pull_request_review_write(method="submit_pending")` |
-| Add PR comment | `mcp_github_add_issue_comment()` |
-| Reply to comment | `mcp_github_add_reply_to_pull_request_comment()` |
-| Scan for secrets | `mcp_github_run_secret_scanning()` |
-| Request Copilot review | `mcp_github_request_copilot_review()` |
+```bash
+source ~/.hermes/skills/github/github-auth/scripts/gh-env.sh
+# Or run the inline setup block from the top of this skill
+```
+
+### Step 2: Gather PR context
+
+Get the PR metadata, description, and list of changed files to understand scope before diving into code.
+
+**With gh:**
+```bash
+gh pr view 123
+gh pr diff 123 --name-only
+gh pr checks 123
+```
+
+**With curl:**
+```bash
+PR_NUMBER=123
+
+# PR details (title, author, description, branch)
+curl -s -H "Authorization: token $GITHUB_TOKEN" \
+  https://api.github.com/repos/$GH_OWNER/$GH_REPO/pulls/$PR_NUMBER
+
+# Changed files with line counts
+curl -s -H "Authorization: token $GITHUB_TOKEN" \
+  https://api.github.com/repos/$GH_OWNER/$GH_REPO/pulls/$PR_NUMBER/files
+```
+
+### Step 3: Check out the PR locally
+
+This gives you full access to `read_file`, `search_files`, and the ability to run tests.
+
+```bash
+git fetch origin pull/$PR_NUMBER/head:pr-$PR_NUMBER
+git checkout pr-$PR_NUMBER
+```
+
+### Step 4: Read the diff and understand changes
+
+```bash
+# Full diff against the base branch
+git diff main...HEAD
+
+# Or file-by-file for large PRs
+git diff main...HEAD --name-only
+# Then for each file:
+git diff main...HEAD -- path/to/file.py
+```
+
+For each changed file, use `read_file` to see full context around the changes — diffs alone can miss issues visible only with surrounding code.
+
+### Step 5: Run automated checks locally (if applicable)
+
+```bash
+# Run tests if there's a test suite
+python -m pytest 2>&1 | tail -20
+# or: npm test, cargo test, go test ./..., etc.
+
+# Run linter if configured
+ruff check . 2>&1 | head -30
+# or: eslint, clippy, etc.
+```
+
+### Step 6: Apply the review checklist (Section 3)
+
+Go through each category: Correctness, Security, Code Quality, Testing, Performance, Documentation.
+
+### Step 7: Post the review to GitHub
+
+Collect your findings and submit them as a formal review with inline comments.
+
+**With gh:**
+```bash
+# If no issues — approve
+gh pr review $PR_NUMBER --approve --body "Reviewed by Hermes Agent. Code looks clean — good test coverage, no security concerns."
+
+# If issues found — request changes with inline comments
+gh pr review $PR_NUMBER --request-changes --body "Found a few issues — see inline comments."
+```
+
+**With curl — atomic review with multiple inline comments:**
+```bash
+HEAD_SHA=$(curl -s -H "Authorization: token $GITHUB_TOKEN" \
+  https://api.github.com/repos/$GH_OWNER/$GH_REPO/pulls/$PR_NUMBER \
+  | python3 -c "import sys,json; print(json.load(sys.stdin)['head']['sha'])")
+
+# Build the review JSON — event is APPROVE, REQUEST_CHANGES, or COMMENT
+curl -s -X POST \
+  -H "Authorization: token $GITHUB_TOKEN" \
+  https://api.github.com/repos/$GH_OWNER/$GH_REPO/pulls/$PR_NUMBER/reviews \
+  -d "{
+    \"commit_id\": \"$HEAD_SHA\",
+    \"event\": \"REQUEST_CHANGES\",
+    \"body\": \"## Hermes Agent Review\n\nFound 2 issues, 1 suggestion. See inline comments.\",
+    \"comments\": [
+      {\"path\": \"src/auth.py\", \"line\": 45, \"body\": \"🔴 **Critical:** User input passed directly to SQL query — use parameterized queries.\"},
+      {\"path\": \"src/models.py\", \"line\": 23, \"body\": \"⚠️ **Warning:** Password stored without hashing.\"},
+      {\"path\": \"src/utils.py\", \"line\": 8, \"body\": \"💡 **Suggestion:** This duplicates logic in core/utils.py:34.\"}
+    ]
+  }"
+```
+
+### Step 8: Also post a summary comment
+
+In addition to inline comments, leave a top-level summary so the PR author gets the full picture at a glance. Use the review output format from `references/review-output-template.md`.
+
+**With gh:**
+```bash
+gh pr comment $PR_NUMBER --body "$(cat <<'EOF'
+## Code Review Summary
+
+**Verdict: Changes Requested** (2 issues, 1 suggestion)
+
+### 🔴 Critical
+- **src/auth.py:45** — SQL injection vulnerability
+
+### ⚠️ Warnings
+- **src/models.py:23** — Plaintext password storage
+
+### 💡 Suggestions
+- **src/utils.py:8** — Duplicated logic, consider consolidating
+
+### ✅ Looks Good
+- Clean API design
+- Good error handling in the middleware layer
+
+---
+*Reviewed by Hermes Agent*
+EOF
+)"
+```
+
+### Step 9: Clean up
+
+```bash
+git checkout main
+git branch -D pr-$PR_NUMBER
+```

 ### Decision: Approve vs Request Changes vs Comment

@@ -1,10 +1,9 @@
 """Tests for agent.auxiliary_client resolution chain, provider overrides, and model overrides."""

 import json
-import logging
 import os
 from pathlib import Path
-from unittest.mock import patch, MagicMock, AsyncMock
+from unittest.mock import patch, MagicMock

 import pytest

@@ -15,7 +14,6 @@ from agent.auxiliary_client import (
    resolve_provider_client,
    auxiliary_max_tokens_param,
    call_llm,
-    async_call_llm,
    _read_codex_access_token,
    _get_auxiliary_provider,
    _get_provider_chain,
@@ -1124,8 +1122,8 @@ class TestCallLlmPaymentFallback:
        exc.status_code = 402
        return exc

-    def test_402_triggers_fallback_when_auto(self, monkeypatch):
-        """When provider is auto and returns 402, call_llm tries the next one."""
+    def test_402_triggers_fallback(self, monkeypatch):
+        """When the primary provider returns 402, call_llm tries the next one."""
        monkeypatch.setenv("OPENROUTER_API_KEY", "or-key")

        primary_client = MagicMock()
@@ -1138,7 +1136,7 @@ class TestCallLlmPaymentFallback:
        with patch("agent.auxiliary_client._get_cached_client",
                    return_value=(primary_client, "google/gemini-3-flash-preview")), \
             patch("agent.auxiliary_client._resolve_task_provider_model",
-                    return_value=("auto", "google/gemini-3-flash-preview", None, None, None)), \
+                    return_value=("openrouter", "google/gemini-3-flash-preview", None, None)), \
             patch("agent.auxiliary_client._try_payment_fallback",
                    return_value=(fallback_client, "gpt-5.2-codex", "openai-codex")) as mock_fb:
            result = call_llm(
@@ -1147,62 +1145,13 @@ class TestCallLlmPaymentFallback:
            )

        assert result is fallback_response
-        mock_fb.assert_called_once_with("auto", "compression", reason="payment error")
+        mock_fb.assert_called_once_with("openrouter", "compression")
        # Fallback call should use the fallback model
        fb_kwargs = fallback_client.chat.completions.create.call_args.kwargs
        assert fb_kwargs["model"] == "gpt-5.2-codex"

-    def test_402_no_fallback_when_explicit_provider(self, monkeypatch):
-        """When provider is explicitly configured (not auto), 402 should NOT fallback (#7559)."""
-        monkeypatch.setenv("OPENROUTER_API_KEY", "or-key")
-
-        primary_client = MagicMock()
-        primary_client.chat.completions.create.side_effect = self._make_402_error()
-
-        with patch("agent.auxiliary_client._get_cached_client",
-                    return_value=(primary_client, "local-model")), \
-             patch("agent.auxiliary_client._resolve_task_provider_model",
-                    return_value=("custom", "local-model", None, None, None)), \
-             patch("agent.auxiliary_client._try_payment_fallback") as mock_fb:
-            with pytest.raises(Exception, match="insufficient credits"):
-                call_llm(
-                    task="compression",
-                    messages=[{"role": "user", "content": "hello"}],
-                )
-
-        # Fallback should NOT be attempted when provider is explicit
-        mock_fb.assert_not_called()
-
-    def test_connection_error_triggers_fallback_when_auto(self, monkeypatch):
-        """Connection errors also trigger fallback when provider is auto."""
-        monkeypatch.setenv("OPENROUTER_API_KEY", "or-key")
-
-        primary_client = MagicMock()
-        conn_err = Exception("Connection refused")
-        conn_err.status_code = None
-        primary_client.chat.completions.create.side_effect = conn_err
-
-        fallback_client = MagicMock()
-        fallback_response = MagicMock()
-        fallback_client.chat.completions.create.return_value = fallback_response
-
-        with patch("agent.auxiliary_client._get_cached_client",
-                    return_value=(primary_client, "model")), \
-             patch("agent.auxiliary_client._resolve_task_provider_model",
-                    return_value=("auto", "model", None, None, None)), \
-             patch("agent.auxiliary_client._is_connection_error", return_value=True), \
-             patch("agent.auxiliary_client._try_payment_fallback",
-                    return_value=(fallback_client, "fb-model", "nous")) as mock_fb:
-            result = call_llm(
-                task="compression",
-                messages=[{"role": "user", "content": "hello"}],
-            )
-
-        assert result is fallback_response
-        mock_fb.assert_called_once_with("auto", "compression", reason="connection error")
-
    def test_non_payment_error_not_caught(self, monkeypatch):
-        """Non-payment/non-connection errors (500) should NOT trigger fallback."""
+        """Non-payment errors (500, connection, etc.) should NOT trigger fallback."""
        monkeypatch.setenv("OPENROUTER_API_KEY", "or-key")

        primary_client = MagicMock()
@@ -1213,7 +1162,7 @@ class TestCallLlmPaymentFallback:
        with patch("agent.auxiliary_client._get_cached_client",
                    return_value=(primary_client, "google/gemini-3-flash-preview")), \
             patch("agent.auxiliary_client._resolve_task_provider_model",
-                    return_value=("auto", "google/gemini-3-flash-preview", None, None, None)):
+                    return_value=("openrouter", "google/gemini-3-flash-preview", None, None)):
            with pytest.raises(Exception, match="Internal Server Error"):
                call_llm(
                    task="compression",
@@ -1230,7 +1179,7 @@ class TestCallLlmPaymentFallback:
        with patch("agent.auxiliary_client._get_cached_client",
                    return_value=(primary_client, "google/gemini-3-flash-preview")), \
             patch("agent.auxiliary_client._resolve_task_provider_model",
-                    return_value=("auto", "google/gemini-3-flash-preview", None, None, None)), \
+                    return_value=("openrouter", "google/gemini-3-flash-preview", None, None)), \
             patch("agent.auxiliary_client._try_payment_fallback",
                    return_value=(None, None, "")):
            with pytest.raises(Exception, match="insufficient credits"):
@@ -1280,283 +1229,3 @@ def test_resolve_api_key_provider_skips_unconfigured_anthropic(monkeypatch):

    assert "anthropic" not in called, \
        "_try_anthropic() should not be called when anthropic is not explicitly configured"
-
-
-# ---------------------------------------------------------------------------
-# model="default" elimination (#7512)
-# ---------------------------------------------------------------------------
-
-
-class TestModelDefaultElimination:
-    """_resolve_api_key_provider must skip providers without known aux models."""
-
-    def test_unknown_provider_skipped(self, monkeypatch):
-        """Providers not in _API_KEY_PROVIDER_AUX_MODELS are skipped, not sent model='default'."""
-        from agent.auxiliary_client import _API_KEY_PROVIDER_AUX_MODELS
-
-        # Verify our known providers have entries
-        assert "gemini" in _API_KEY_PROVIDER_AUX_MODELS
-        assert "kimi-coding" in _API_KEY_PROVIDER_AUX_MODELS
-
-        # A random provider_id not in the dict should return None
-        assert _API_KEY_PROVIDER_AUX_MODELS.get("totally-unknown-provider") is None
-
-    def test_known_provider_gets_real_model(self):
-        """Known providers get a real model name, not 'default'."""
-        from agent.auxiliary_client import _API_KEY_PROVIDER_AUX_MODELS
-
-        for provider_id, model in _API_KEY_PROVIDER_AUX_MODELS.items():
-            assert model != "default", f"{provider_id} should not map to 'default'"
-            assert isinstance(model, str) and model.strip(), \
-                f"{provider_id} should have a non-empty model string"
-
-
-# ---------------------------------------------------------------------------
-# _try_payment_fallback reason parameter (#7512 bug 3)
-# ---------------------------------------------------------------------------
-
-
-class TestTryPaymentFallbackReason:
-    """_try_payment_fallback uses the reason parameter in log messages."""
-
-    def test_reason_parameter_passed_through(self, monkeypatch):
-        """The reason= parameter is accepted without error."""
-        from agent.auxiliary_client import _try_payment_fallback
-
-        # Mock the provider chain to return nothing
-        monkeypatch.setattr(
-            "agent.auxiliary_client._get_provider_chain",
-            lambda: [],
-        )
-        monkeypatch.setattr(
-            "agent.auxiliary_client._read_main_provider",
-            lambda: "",
-        )
-
-        client, model, label = _try_payment_fallback(
-            "openrouter", task="compression", reason="connection error"
-        )
-        assert client is None
-        assert label == ""
-
-
-# ---------------------------------------------------------------------------
-# _is_connection_error coverage
-# ---------------------------------------------------------------------------
-
-
-class TestIsConnectionError:
-    """Tests for _is_connection_error detection."""
-
-    def test_connection_refused(self):
-        from agent.auxiliary_client import _is_connection_error
-        err = Exception("Connection refused")
-        assert _is_connection_error(err) is True
-
-    def test_timeout(self):
-        from agent.auxiliary_client import _is_connection_error
-        err = Exception("Request timed out.")
-        assert _is_connection_error(err) is True
-
-    def test_dns_failure(self):
-        from agent.auxiliary_client import _is_connection_error
-        err = Exception("Name or service not known")
-        assert _is_connection_error(err) is True
-
-    def test_normal_api_error_not_connection(self):
-        from agent.auxiliary_client import _is_connection_error
-        err = Exception("Bad Request: invalid model")
-        err.status_code = 400
-        assert _is_connection_error(err) is False
-
-    def test_500_not_connection(self):
-        from agent.auxiliary_client import _is_connection_error
-        err = Exception("Internal Server Error")
-        err.status_code = 500
-        assert _is_connection_error(err) is False
-
-
-# ---------------------------------------------------------------------------
-# async_call_llm payment / connection fallback (#7512 bug 2)
-# ---------------------------------------------------------------------------
-
-
-class TestAsyncCallLlmFallback:
-    """async_call_llm mirrors call_llm fallback behavior."""
-
-    def _make_402_error(self, msg="Payment Required: insufficient credits"):
-        exc = Exception(msg)
-        exc.status_code = 402
-        return exc
-
-    @pytest.mark.asyncio
-    async def test_402_triggers_async_fallback_when_auto(self, monkeypatch):
-        """When provider is auto and returns 402, async_call_llm tries fallback."""
-        monkeypatch.setenv("OPENROUTER_API_KEY", "or-key")
-
-        primary_client = MagicMock()
-        primary_client.chat.completions.create = AsyncMock(
-            side_effect=self._make_402_error())
-
-        # Fallback client (sync) returned by _try_payment_fallback
-        fb_sync_client = MagicMock()
-        fb_async_client = MagicMock()
-        fb_response = MagicMock()
-        fb_async_client.chat.completions.create = AsyncMock(return_value=fb_response)
-
-        with patch("agent.auxiliary_client._get_cached_client",
-                    return_value=(primary_client, "google/gemini-3-flash-preview")), \
-             patch("agent.auxiliary_client._resolve_task_provider_model",
-                    return_value=("auto", "google/gemini-3-flash-preview", None, None, None)), \
-             patch("agent.auxiliary_client._try_payment_fallback",
-                    return_value=(fb_sync_client, "gpt-5.2-codex", "openai-codex")) as mock_fb, \
-             patch("agent.auxiliary_client._to_async_client",
-                    return_value=(fb_async_client, "gpt-5.2-codex")):
-            result = await async_call_llm(
-                task="compression",
-                messages=[{"role": "user", "content": "hello"}],
-            )
-
-        assert result is fb_response
-        mock_fb.assert_called_once_with("auto", "compression", reason="payment error")
-
-    @pytest.mark.asyncio
-    async def test_402_no_async_fallback_when_explicit(self, monkeypatch):
-        """When provider is explicit, 402 should NOT trigger async fallback."""
-        monkeypatch.setenv("OPENROUTER_API_KEY", "or-key")
-
-        primary_client = MagicMock()
-        primary_client.chat.completions.create = AsyncMock(
-            side_effect=self._make_402_error())
-
-        with patch("agent.auxiliary_client._get_cached_client",
-                    return_value=(primary_client, "local-model")), \
-             patch("agent.auxiliary_client._resolve_task_provider_model",
-                    return_value=("custom", "local-model", None, None, None)), \
-             patch("agent.auxiliary_client._try_payment_fallback") as mock_fb:
-            with pytest.raises(Exception, match="insufficient credits"):
-                await async_call_llm(
-                    task="compression",
-                    messages=[{"role": "user", "content": "hello"}],
-                )
-
-        mock_fb.assert_not_called()
-
-    @pytest.mark.asyncio
-    async def test_connection_error_triggers_async_fallback(self, monkeypatch):
-        """Connection errors trigger async fallback when provider is auto."""
-        monkeypatch.setenv("OPENROUTER_API_KEY", "or-key")
-
-        primary_client = MagicMock()
-        conn_err = Exception("Connection refused")
-        conn_err.status_code = None
-        primary_client.chat.completions.create = AsyncMock(side_effect=conn_err)
-
-        fb_sync_client = MagicMock()
-        fb_async_client = MagicMock()
-        fb_response = MagicMock()
-        fb_async_client.chat.completions.create = AsyncMock(return_value=fb_response)
-
-        with patch("agent.auxiliary_client._get_cached_client",
-                    return_value=(primary_client, "model")), \
-             patch("agent.auxiliary_client._resolve_task_provider_model",
-                    return_value=("auto", "model", None, None, None)), \
-             patch("agent.auxiliary_client._is_connection_error", return_value=True), \
-             patch("agent.auxiliary_client._try_payment_fallback",
-                    return_value=(fb_sync_client, "fb-model", "nous")) as mock_fb, \
-             patch("agent.auxiliary_client._to_async_client",
-                    return_value=(fb_async_client, "fb-model")):
-            result = await async_call_llm(
-                task="compression",
-                messages=[{"role": "user", "content": "hello"}],
-            )
-
-        assert result is fb_response
-        mock_fb.assert_called_once_with("auto", "compression", reason="connection error")
-class TestStaleBaseUrlWarning:
-    """_resolve_auto() warns when OPENAI_BASE_URL conflicts with config provider (#5161)."""
-
-    def test_warns_when_openai_base_url_set_with_named_provider(self, monkeypatch, caplog):
-        """Warning fires when OPENAI_BASE_URL is set but provider is a named provider."""
-        import agent.auxiliary_client as mod
-        # Reset the module-level flag so the warning fires
-        monkeypatch.setattr(mod, "_stale_base_url_warned", False)
-        monkeypatch.setenv("OPENAI_BASE_URL", "http://localhost:11434/v1")
-        monkeypatch.setenv("OPENROUTER_API_KEY", "sk-or-test")
-
-        with patch("agent.auxiliary_client._read_main_provider", return_value="openrouter"), \
-             patch("agent.auxiliary_client._read_main_model", return_value="google/gemini-flash"), \
-             caplog.at_level(logging.WARNING, logger="agent.auxiliary_client"):
-            _resolve_auto()
-
-        assert any("OPENAI_BASE_URL is set" in rec.message for rec in caplog.records), \
-            "Expected a warning about stale OPENAI_BASE_URL"
-        assert mod._stale_base_url_warned is True
-
-    def test_no_warning_when_provider_is_custom(self, monkeypatch, caplog):
-        """No warning when the provider is 'custom' — OPENAI_BASE_URL is expected."""
-        import agent.auxiliary_client as mod
-        monkeypatch.setattr(mod, "_stale_base_url_warned", False)
-        monkeypatch.setenv("OPENAI_BASE_URL", "http://localhost:11434/v1")
-        monkeypatch.setenv("OPENAI_API_KEY", "test-key")
-
-        with patch("agent.auxiliary_client._read_main_provider", return_value="custom"), \
-             patch("agent.auxiliary_client._read_main_model", return_value="llama3"), \
-             patch("agent.auxiliary_client._resolve_custom_runtime",
-                   return_value=("http://localhost:11434/v1", "test-key", None)), \
-             patch("agent.auxiliary_client.OpenAI") as mock_openai, \
-             caplog.at_level(logging.WARNING, logger="agent.auxiliary_client"):
-            mock_openai.return_value = MagicMock()
-            _resolve_auto()
-
-        assert not any("OPENAI_BASE_URL is set" in rec.message for rec in caplog.records), \
-            "Should NOT warn when provider is 'custom'"
-
-    def test_no_warning_when_provider_is_named_custom(self, monkeypatch, caplog):
-        """No warning when the provider is 'custom:myname' — base_url comes from config."""
-        import agent.auxiliary_client as mod
-        monkeypatch.setattr(mod, "_stale_base_url_warned", False)
-        monkeypatch.setenv("OPENAI_BASE_URL", "http://localhost:11434/v1")
-        monkeypatch.setenv("OPENAI_API_KEY", "test-key")
-
-        with patch("agent.auxiliary_client._read_main_provider", return_value="custom:ollama-local"), \
-             patch("agent.auxiliary_client._read_main_model", return_value="llama3"), \
-             patch("agent.auxiliary_client.resolve_provider_client",
-                   return_value=(MagicMock(), "llama3")), \
-             caplog.at_level(logging.WARNING, logger="agent.auxiliary_client"):
-            _resolve_auto()
-
-        assert not any("OPENAI_BASE_URL is set" in rec.message for rec in caplog.records), \
-            "Should NOT warn when provider is 'custom:*'"
-
-    def test_no_warning_when_openai_base_url_not_set(self, monkeypatch, caplog):
-        """No warning when OPENAI_BASE_URL is absent."""
-        import agent.auxiliary_client as mod
-        monkeypatch.setattr(mod, "_stale_base_url_warned", False)
-        monkeypatch.delenv("OPENAI_BASE_URL", raising=False)
-        monkeypatch.setenv("OPENROUTER_API_KEY", "sk-or-test")
-
-        with patch("agent.auxiliary_client._read_main_provider", return_value="openrouter"), \
-             patch("agent.auxiliary_client._read_main_model", return_value="google/gemini-flash"), \
-             caplog.at_level(logging.WARNING, logger="agent.auxiliary_client"):
-            _resolve_auto()
-
-        assert not any("OPENAI_BASE_URL is set" in rec.message for rec in caplog.records), \
-            "Should NOT warn when OPENAI_BASE_URL is not set"
-
-    def test_warning_only_fires_once(self, monkeypatch, caplog):
-        """Warning is suppressed after the first invocation."""
-        import agent.auxiliary_client as mod
-        monkeypatch.setattr(mod, "_stale_base_url_warned", False)
-        monkeypatch.setenv("OPENAI_BASE_URL", "http://localhost:11434/v1")
-        monkeypatch.setenv("OPENROUTER_API_KEY", "sk-or-test")
-
-        with patch("agent.auxiliary_client._read_main_provider", return_value="openrouter"), \
-             patch("agent.auxiliary_client._read_main_model", return_value="google/gemini-flash"), \
-             caplog.at_level(logging.WARNING, logger="agent.auxiliary_client"):
-            _resolve_auto()
-            caplog.clear()
-            _resolve_auto()
-
-        assert not any("OPENAI_BASE_URL is set" in rec.message for rec in caplog.records), \
-            "Warning should not fire a second time"
@@ -1,139 +0,0 @@
-"""Tests for focus_topic flowing through the compressor.
-
-Verifies that _generate_summary and compress accept and use the focus_topic
-parameter correctly.  Inspired by Claude Code's /compact <focus>.
-"""
-
-from unittest.mock import MagicMock, patch
-
-from agent.context_compressor import ContextCompressor
-
-
-def _make_compressor():
-    """Create a ContextCompressor with minimal state for testing."""
-    compressor = ContextCompressor.__new__(ContextCompressor)
-    compressor.protect_first_n = 2
-    compressor.protect_last_n = 5
-    compressor.tail_token_budget = 20000
-    compressor.context_length = 200000
-    compressor.threshold_percent = 0.80
-    compressor.threshold_tokens = 160000
-    compressor.max_summary_tokens = 10000
-    compressor.quiet_mode = True
-    compressor.compression_count = 0
-    compressor.last_prompt_tokens = 0
-    compressor._previous_summary = None
-    compressor._summary_failure_cooldown_until = 0.0
-    compressor.summary_model = None
-    return compressor
-
-
-def test_focus_topic_injected_into_summary_prompt():
-    """When focus_topic is provided, the LLM prompt includes focus guidance."""
-    compressor = _make_compressor()
-    turns = [
-        {"role": "user", "content": "Tell me about the database schema"},
-        {"role": "assistant", "content": "The schema has tables: users, orders, products."},
-    ]
-
-    captured_prompt = {}
-
-    def mock_call_llm(**kwargs):
-        captured_prompt["messages"] = kwargs["messages"]
-        resp = MagicMock()
-        resp.choices = [MagicMock()]
-        resp.choices[0].message.content = "## Goal\nUnderstand DB schema."
-        return resp
-
-    with patch("agent.context_compressor.call_llm", mock_call_llm):
-        result = compressor._generate_summary(turns, focus_topic="database schema")
-
-    assert result is not None
-    prompt_text = captured_prompt["messages"][0]["content"]
-    assert 'FOCUS TOPIC: "database schema"' in prompt_text
-    assert "PRIORITISE" in prompt_text
-    assert "60-70%" in prompt_text
-
-
-def test_no_focus_topic_no_injection():
-    """Without focus_topic, the prompt doesn't contain focus guidance."""
-    compressor = _make_compressor()
-    turns = [
-        {"role": "user", "content": "Hello"},
-        {"role": "assistant", "content": "Hi"},
-    ]
-
-    captured_prompt = {}
-
-    def mock_call_llm(**kwargs):
-        captured_prompt["messages"] = kwargs["messages"]
-        resp = MagicMock()
-        resp.choices = [MagicMock()]
-        resp.choices[0].message.content = "## Goal\nGreeting."
-        return resp
-
-    with patch("agent.context_compressor.call_llm", mock_call_llm):
-        result = compressor._generate_summary(turns)
-
-    prompt_text = captured_prompt["messages"][0]["content"]
-    assert "FOCUS TOPIC" not in prompt_text
-
-
-def test_compress_passes_focus_to_generate_summary():
-    """compress() passes focus_topic through to _generate_summary."""
-    compressor = _make_compressor()
-
-    # Track what _generate_summary receives
-    received_kwargs = {}
-    original_generate = compressor._generate_summary
-
-    def tracking_generate(turns, **kwargs):
-        received_kwargs.update(kwargs)
-        return "## Goal\nTest."
-
-    compressor._generate_summary = tracking_generate
-
-    messages = [
-        {"role": "system", "content": "System prompt"},
-        {"role": "user", "content": "first"},
-        {"role": "assistant", "content": "reply1"},
-        {"role": "user", "content": "second"},
-        {"role": "assistant", "content": "reply2"},
-        {"role": "user", "content": "third"},
-        {"role": "assistant", "content": "reply3"},
-        {"role": "user", "content": "fourth"},
-        {"role": "assistant", "content": "reply4"},
-    ]
-
-    compressor.compress(messages, current_tokens=100000, focus_topic="authentication flow")
-
-    assert received_kwargs.get("focus_topic") == "authentication flow"
-
-
-def test_compress_none_focus_by_default():
-    """compress() passes None focus_topic by default."""
-    compressor = _make_compressor()
-
-    received_kwargs = {}
-
-    def tracking_generate(turns, **kwargs):
-        received_kwargs.update(kwargs)
-        return "## Goal\nTest."
-
-    compressor._generate_summary = tracking_generate
-
-    messages = [
-        {"role": "system", "content": "System prompt"},
-        {"role": "user", "content": "first"},
-        {"role": "assistant", "content": "reply1"},
-        {"role": "user", "content": "second"},
-        {"role": "assistant", "content": "reply2"},
-        {"role": "user", "content": "third"},
-        {"role": "assistant", "content": "reply3"},
-        {"role": "user", "content": "fourth"},
-        {"role": "assistant", "content": "reply4"},
-    ]
-
-    compressor.compress(messages, current_tokens=100000)
-
-    assert received_kwargs.get("focus_topic") is None
@@ -576,19 +576,11 @@ class TestSummaryTargetRatio:
        assert c.summary_target_ratio == 0.80

    def test_default_threshold_is_50_percent(self):
-        """Default compression threshold should be 50%, with a 64K floor."""
+        """Default compression threshold should be 50%."""
        with patch("agent.context_compressor.get_model_context_length", return_value=100_000):
            c = ContextCompressor(model="test", quiet_mode=True)
        assert c.threshold_percent == 0.50
-        # 50% of 100K = 50K, but the floor is 64K
-        assert c.threshold_tokens == 64_000
-
-    def test_threshold_floor_does_not_apply_above_128k(self):
-        """On large-context models the 50% percentage is used directly."""
-        with patch("agent.context_compressor.get_model_context_length", return_value=200_000):
-            c = ContextCompressor(model="test", quiet_mode=True)
-        # 50% of 200K = 100K, which is above the 64K floor
-        assert c.threshold_tokens == 100_000
+        assert c.threshold_tokens == 50_000

    def test_default_protect_last_n_is_20(self):
        """Default protect_last_n should be 20."""
@@ -22,9 +22,6 @@ class TestLocalStreamReadTimeout:
        "http://0.0.0.0:5000",
        "http://192.168.1.100:8000",
        "http://10.0.0.5:1234",
-        "http://host.docker.internal:11434",
-        "http://host.containers.internal:11434",
-        "http://host.lima.internal:11434",
    ])
    def test_local_endpoint_bumps_read_timeout(self, base_url):
        """Local endpoint + default timeout -> bumps to base_timeout."""
@@ -71,38 +68,3 @@ class TestLocalStreamReadTimeout:
            if _stream_read_timeout == 120.0 and base_url and is_local_endpoint(base_url):
                _stream_read_timeout = _base_timeout
            assert _stream_read_timeout == 120.0
-
-
-class TestIsLocalEndpoint:
-    """Direct unit tests for is_local_endpoint."""
-
-    @pytest.mark.parametrize("url", [
-        "http://localhost:11434",
-        "http://127.0.0.1:8080",
-        "http://0.0.0.0:5000",
-        "http://[::1]:11434",
-        "http://192.168.1.100:8000",
-        "http://10.0.0.5:1234",
-        "http://172.17.0.1:11434",
-    ])
-    def test_classic_local_addresses(self, url):
-        assert is_local_endpoint(url) is True
-
-    @pytest.mark.parametrize("url", [
-        "http://host.docker.internal:11434",
-        "http://host.docker.internal:8080/v1",
-        "http://gateway.docker.internal:11434",
-        "http://host.containers.internal:11434",
-        "http://host.lima.internal:11434",
-    ])
-    def test_container_dns_names(self, url):
-        assert is_local_endpoint(url) is True
-
-    @pytest.mark.parametrize("url", [
-        "https://api.openai.com",
-        "https://openrouter.ai/api",
-        "https://api.anthropic.com",
-        "https://evil.docker.internal.example.com",
-    ])
-    def test_remote_endpoints(self, url):
-        assert is_local_endpoint(url) is False
@@ -1,37 +1,37 @@
-"""Tests for MiniMax provider hardening — context lengths, thinking, catalog, beta headers, transport."""
+"""Tests for MiniMax provider hardening — context lengths, thinking guard, catalog, beta headers."""

 from unittest.mock import patch


 class TestMinimaxContextLengths:
-    """Verify context length entries match official docs (204,800 for all models).
+    """Verify per-model context length entries for MiniMax models."""

-    Source: https://platform.minimax.io/docs/api-reference/text-anthropic-api
-    """
-
-    def test_minimax_prefix_has_correct_context(self):
+    def test_m1_variants_have_1m_context(self):
        from agent.model_metadata import DEFAULT_CONTEXT_LENGTHS
-        assert DEFAULT_CONTEXT_LENGTHS["minimax"] == 204_800
+        # Keys are lowercase because the lookup lowercases model names
+        for model in ("minimax-m1", "minimax-m1-40k", "minimax-m1-80k",
+                       "minimax-m1-128k", "minimax-m1-256k"):
+            assert model in DEFAULT_CONTEXT_LENGTHS, f"{model} missing from context lengths"
+            assert DEFAULT_CONTEXT_LENGTHS[model] == 1_000_000, f"{model} expected 1M"

-    def test_minimax_models_resolve_via_prefix(self):
-        from agent.model_metadata import get_model_context_length
-        # All MiniMax models should resolve to 204,800 via the "minimax" prefix
-        for model in ("MiniMax-M2.7", "MiniMax-M2.5", "MiniMax-M2.1", "MiniMax-M2"):
-            ctx = get_model_context_length(model, "")
-            assert ctx == 204_800, f"{model} expected 204800, got {ctx}"
+    def test_m2_variants_have_1m_context(self):
+        from agent.model_metadata import DEFAULT_CONTEXT_LENGTHS
+        # Keys are lowercase because the lookup lowercases model names
+        for model in ("minimax-m2.5", "minimax-m2.7"):
+            assert model in DEFAULT_CONTEXT_LENGTHS, f"{model} missing from context lengths"
+            assert DEFAULT_CONTEXT_LENGTHS[model] == 1_048_576, f"{model} expected 1048576"
+
+    def test_minimax_prefix_fallback(self):
+        from agent.model_metadata import DEFAULT_CONTEXT_LENGTHS
+        # The generic "minimax" prefix entry should be 1M for unknown models
+        assert DEFAULT_CONTEXT_LENGTHS["minimax"] == 1_048_576



-class TestMinimaxThinkingSupport:
-    """Verify that MiniMax gets manual thinking (not adaptive).
+class TestMinimaxThinkingGuard:
+    """Verify that build_anthropic_kwargs does NOT add thinking params for MiniMax models."""

-    MiniMax's Anthropic-compat endpoint officially supports the thinking
-    parameter (https://platform.minimax.io/docs/api-reference/text-anthropic-api).
-    It should get manual thinking (type=enabled + budget_tokens), NOT adaptive
-    thinking (which is Claude 4.6-only).
-    """
-
-    def test_minimax_m27_gets_manual_thinking(self):
+    def test_no_thinking_for_minimax_m27(self):
        from agent.anthropic_adapter import build_anthropic_kwargs
        kwargs = build_anthropic_kwargs(
            model="MiniMax-M2.7",
@@ -40,23 +40,19 @@ class TestMinimaxThinkingSupport:
            max_tokens=4096,
            reasoning_config={"enabled": True, "effort": "medium"},
        )
-        assert "thinking" in kwargs
-        assert kwargs["thinking"]["type"] == "enabled"
-        assert "budget_tokens" in kwargs["thinking"]
-        # MiniMax should NOT get adaptive thinking or output_config
+        assert "thinking" not in kwargs
        assert "output_config" not in kwargs

-    def test_minimax_m25_gets_manual_thinking(self):
+    def test_no_thinking_for_minimax_m1(self):
        from agent.anthropic_adapter import build_anthropic_kwargs
        kwargs = build_anthropic_kwargs(
-            model="MiniMax-M2.5",
+            model="MiniMax-M1-128k",
            messages=[{"role": "user", "content": "hello"}],
            tools=None,
            max_tokens=4096,
            reasoning_config={"enabled": True, "effort": "high"},
        )
-        assert "thinking" in kwargs
-        assert kwargs["thinking"]["type"] == "enabled"
+        assert "thinking" not in kwargs

    def test_thinking_still_works_for_claude(self):
        from agent.anthropic_adapter import build_anthropic_kwargs
@@ -85,30 +81,25 @@ class TestMinimaxAuxModel:


 class TestMinimaxModelCatalog:
-    """Verify the model catalog matches official Anthropic-compat endpoint models.
+    """Verify the model catalog includes M1 family and excludes deprecated models."""

-    Source: https://platform.minimax.io/docs/api-reference/text-anthropic-api
-    """
-
-    def test_catalog_includes_current_models(self):
+    def test_catalog_includes_m1_family(self):
        from hermes_cli.models import _PROVIDER_MODELS
        for provider in ("minimax", "minimax-cn"):
            models = _PROVIDER_MODELS[provider]
-            assert "MiniMax-M2.7" in models
-            assert "MiniMax-M2.5" in models
-            assert "MiniMax-M2.1" in models
-            assert "MiniMax-M2" in models
+            assert "MiniMax-M1" in models
+            assert "MiniMax-M1-40k" in models
+            assert "MiniMax-M1-80k" in models
+            assert "MiniMax-M1-128k" in models
+            assert "MiniMax-M1-256k" in models

-    def test_catalog_excludes_m1_family(self):
-        """M1 models are not available on the /anthropic endpoint."""
+    def test_catalog_excludes_deprecated(self):
        from hermes_cli.models import _PROVIDER_MODELS
        for provider in ("minimax", "minimax-cn"):
            models = _PROVIDER_MODELS[provider]
-            assert "MiniMax-M1" not in models
+            assert "MiniMax-M2.1" not in models

    def test_catalog_excludes_highspeed(self):
-        """Highspeed variants are available but not shown in default catalog
-        (users can still specify them manually)."""
        from hermes_cli.models import _PROVIDER_MODELS
        for provider in ("minimax", "minimax-cn"):
            models = _PROVIDER_MODELS[provider]
@@ -211,154 +202,3 @@ class TestMinimaxBetaHeaders:
    def test_common_betas_regular_url(self):
        from agent.anthropic_adapter import _common_betas_for_base_url, _COMMON_BETAS
        assert _common_betas_for_base_url("https://api.anthropic.com") == _COMMON_BETAS
-
-
-class TestMinimaxApiMode:
-    """Verify determine_api_mode returns anthropic_messages for MiniMax providers.
-
-    The MiniMax /anthropic endpoint speaks Anthropic Messages wire format,
-    not OpenAI chat completions.  The overlay transport must reflect this
-    so that code paths calling determine_api_mode() without a base_url
-    (e.g. /model switch) get the correct api_mode.
-    """
-
-    def test_minimax_returns_anthropic_messages(self):
-        from hermes_cli.providers import determine_api_mode
-        assert determine_api_mode("minimax") == "anthropic_messages"
-
-    def test_minimax_cn_returns_anthropic_messages(self):
-        from hermes_cli.providers import determine_api_mode
-        assert determine_api_mode("minimax-cn") == "anthropic_messages"
-
-    def test_minimax_with_url_also_works(self):
-        from hermes_cli.providers import determine_api_mode
-        # Even with explicit base_url, provider lookup takes priority
-        assert determine_api_mode("minimax", "https://api.minimax.io/anthropic") == "anthropic_messages"
-
-    def test_anthropic_still_returns_anthropic_messages(self):
-        from hermes_cli.providers import determine_api_mode
-        assert determine_api_mode("anthropic") == "anthropic_messages"
-
-    def test_openai_returns_chat_completions(self):
-        from hermes_cli.providers import determine_api_mode
-        # Sanity check: standard providers are unaffected
-        result = determine_api_mode("deepseek")
-        assert result == "chat_completions"
-
-
-class TestMinimaxMaxOutput:
-    """Verify _get_anthropic_max_output returns correct limits for MiniMax models.
-
-    MiniMax max output is 131,072 tokens (source: OpenClaw model definitions,
-    cross-referenced with MiniMax API behavior).
-    """
-
-    def test_minimax_m27_output_limit(self):
-        from agent.anthropic_adapter import _get_anthropic_max_output
-        assert _get_anthropic_max_output("MiniMax-M2.7") == 131_072
-
-    def test_minimax_m25_output_limit(self):
-        from agent.anthropic_adapter import _get_anthropic_max_output
-        assert _get_anthropic_max_output("MiniMax-M2.5") == 131_072
-
-    def test_minimax_m2_output_limit(self):
-        from agent.anthropic_adapter import _get_anthropic_max_output
-        assert _get_anthropic_max_output("MiniMax-M2") == 131_072
-
-    def test_claude_output_unaffected(self):
-        from agent.anthropic_adapter import _get_anthropic_max_output
-        # Sanity: Claude limits are not broken by the MiniMax entry
-        assert _get_anthropic_max_output("claude-sonnet-4-6") == 64_000
-
-
-class TestMinimaxPreserveDots:
-    """Verify that MiniMax model names preserve dots through the Anthropic adapter.
-
-    MiniMax model IDs like 'MiniMax-M2.7' must NOT have dots converted to
-    hyphens — the endpoint expects the exact name with dots.
-    """
-
-    def test_minimax_provider_preserves_dots(self):
-        from types import SimpleNamespace
-        agent = SimpleNamespace(provider="minimax", base_url="")
-        from run_agent import AIAgent
-        assert AIAgent._anthropic_preserve_dots(agent) is True
-
-    def test_minimax_cn_provider_preserves_dots(self):
-        from types import SimpleNamespace
-        agent = SimpleNamespace(provider="minimax-cn", base_url="")
-        from run_agent import AIAgent
-        assert AIAgent._anthropic_preserve_dots(agent) is True
-
-    def test_minimax_url_preserves_dots(self):
-        from types import SimpleNamespace
-        agent = SimpleNamespace(provider="custom", base_url="https://api.minimax.io/anthropic")
-        from run_agent import AIAgent
-        assert AIAgent._anthropic_preserve_dots(agent) is True
-
-    def test_minimax_cn_url_preserves_dots(self):
-        from types import SimpleNamespace
-        agent = SimpleNamespace(provider="custom", base_url="https://api.minimaxi.com/anthropic")
-        from run_agent import AIAgent
-        assert AIAgent._anthropic_preserve_dots(agent) is True
-
-    def test_anthropic_does_not_preserve_dots(self):
-        from types import SimpleNamespace
-        agent = SimpleNamespace(provider="anthropic", base_url="https://api.anthropic.com")
-        from run_agent import AIAgent
-        assert AIAgent._anthropic_preserve_dots(agent) is False
-
-    def test_normalize_preserves_m27_dot(self):
-        from agent.anthropic_adapter import normalize_model_name
-        assert normalize_model_name("MiniMax-M2.7", preserve_dots=True) == "MiniMax-M2.7"
-
-    def test_normalize_converts_without_preserve(self):
-        from agent.anthropic_adapter import normalize_model_name
-        # Without preserve_dots, dots become hyphens (broken for MiniMax)
-        assert normalize_model_name("MiniMax-M2.7", preserve_dots=False) == "MiniMax-M2-7"
-
-
-class TestMinimaxSwitchModelCredentialGuard:
-    """Verify switch_model() does not leak Anthropic credentials to MiniMax.
-
-    The __init__ path correctly guards against this (line 761), but switch_model()
-    must mirror that guard. Without it, /model switch to minimax with no explicit
-    api_key would fall back to resolve_anthropic_token() and send Anthropic creds
-    to the MiniMax endpoint.
-    """
-
-    def test_switch_to_minimax_does_not_resolve_anthropic_token(self):
-        """switch_model() should NOT call resolve_anthropic_token() for MiniMax."""
-        from unittest.mock import patch, MagicMock
-
-        with patch("run_agent.AIAgent.__init__", return_value=None):
-            from run_agent import AIAgent
-            agent = AIAgent.__new__(AIAgent)
-            agent.provider = "anthropic"
-            agent.model = "claude-sonnet-4"
-            agent.api_key = "sk-ant-fake"
-            agent.base_url = "https://api.anthropic.com"
-            agent.api_mode = "anthropic_messages"
-            agent._anthropic_base_url = "https://api.anthropic.com"
-            agent._anthropic_api_key = "sk-ant-fake"
-            agent._is_anthropic_oauth = False
-            agent._client_kwargs = {}
-            agent.client = None
-            agent._anthropic_client = MagicMock()
-
-        with patch("agent.anthropic_adapter.build_anthropic_client") as mock_build, \
-             patch("agent.anthropic_adapter.resolve_anthropic_token", return_value="sk-ant-leaked") as mock_resolve, \
-             patch("agent.anthropic_adapter._is_oauth_token", return_value=False):
-
-            agent.switch_model(
-                new_model="MiniMax-M2.7",
-                new_provider="minimax",
-                api_mode="anthropic_messages",
-                api_key="mm-key-123",
-                base_url="https://api.minimax.io/anthropic",
-            )
-            # resolve_anthropic_token should NOT be called for non-Anthropic providers
-            mock_resolve.assert_not_called()
-            # The key passed to build_anthropic_client should be the MiniMax key
-            build_args = mock_build.call_args
-            assert build_args[0][0] == "mm-key-123"
@@ -50,8 +50,7 @@ class TestEstimateTokensRough:
        assert estimate_tokens_rough("a" * 400) == 100

    def test_short_text(self):
-        # "hello" = 5 chars → ceil(5/4) = 2
-        assert estimate_tokens_rough("hello") == 2
+        assert estimate_tokens_rough("hello") == 1

    def test_proportional(self):
        short = estimate_tokens_rough("hello world")
@@ -69,11 +68,10 @@ class TestEstimateMessagesTokensRough:
        assert estimate_messages_tokens_rough([]) == 0

    def test_single_message_concrete_value(self):
-        """Verify against known str(msg) length (ceiling division)."""
+        """Verify against known str(msg) length."""
        msg = {"role": "user", "content": "a" * 400}
        result = estimate_messages_tokens_rough([msg])
-        n = len(str(msg))
-        expected = (n + 3) // 4
+        expected = len(str(msg)) // 4
        assert result == expected

    def test_multiple_messages_additive(self):
@@ -82,8 +80,7 @@ class TestEstimateMessagesTokensRough:
            {"role": "assistant", "content": "Hi there, how can I help?"},
        ]
        result = estimate_messages_tokens_rough(msgs)
-        n = sum(len(str(m)) for m in msgs)
-        expected = (n + 3) // 4
+        expected = sum(len(str(m)) for m in msgs) // 4
        assert result == expected

    def test_tool_call_message(self):
@@ -92,7 +89,7 @@ class TestEstimateMessagesTokensRough:
               "tool_calls": [{"id": "1", "function": {"name": "terminal", "arguments": "{}"}}]}
        result = estimate_messages_tokens_rough([msg])
        assert result > 0
-        assert result == (len(str(msg)) + 3) // 4
+        assert result == len(str(msg)) // 4

    def test_message_with_list_content(self):
        """Vision messages with multimodal content arrays."""
@@ -101,7 +98,7 @@ class TestEstimateMessagesTokensRough:
            {"type": "image_url", "image_url": {"url": "data:image/png;base64,AAAA"}}
        ]}
        result = estimate_messages_tokens_rough([msg])
-        assert result == (len(str(msg)) + 3) // 4
+        assert result == len(str(msg)) // 4


 # =========================================================================
@@ -225,24 +222,6 @@ class TestGetModelContextLength:
        mock_fetch.return_value = {}
        assert get_model_context_length("openai/gpt-4o") == 128000

-    @patch("agent.model_metadata.fetch_model_metadata")
-    def test_qwen3_coder_plus_context_length(self, mock_fetch):
-        """qwen3-coder-plus has a 1M context window, not the generic 128K Qwen default."""
-        mock_fetch.return_value = {}
-        assert get_model_context_length("qwen3-coder-plus") == 1000000
-
-    @patch("agent.model_metadata.fetch_model_metadata")
-    def test_qwen3_coder_context_length(self, mock_fetch):
-        """qwen3-coder has a 256K context window, not the generic 128K Qwen default."""
-        mock_fetch.return_value = {}
-        assert get_model_context_length("qwen3-coder") == 262144
-
-    @patch("agent.model_metadata.fetch_model_metadata")
-    def test_qwen_generic_context_length(self, mock_fetch):
-        """Generic qwen models still get the 128K default."""
-        mock_fetch.return_value = {}
-        assert get_model_context_length("qwen3-plus") == 131072
-
    @patch("agent.model_metadata.fetch_model_metadata")
    def test_api_missing_context_length_key(self, mock_fetch):
        """Model in API but without context_length → defaults to 128000."""
@@ -7,7 +7,6 @@ from agent.models_dev import (
    PROVIDER_TO_MODELS_DEV,
    _extract_context,
    fetch_models_dev,
-    get_model_capabilities,
    lookup_models_dev_context,
 )

@@ -196,88 +195,3 @@ class TestFetchModelsDev:
        result = fetch_models_dev()
        mock_get.assert_not_called()
        assert result == SAMPLE_REGISTRY
-
-
-# ---------------------------------------------------------------------------
-# get_model_capabilities — vision via modalities.input
-# ---------------------------------------------------------------------------
-
-
-CAPS_REGISTRY = {
-    "google": {
-        "id": "google",
-        "models": {
-            "gemma-4-31b-it": {
-                "id": "gemma-4-31b-it",
-                "attachment": False,
-                "tool_call": True,
-                "modalities": {"input": ["text", "image"]},
-                "limit": {"context": 128000, "output": 8192},
-            },
-            "gemma-3-1b": {
-                "id": "gemma-3-1b",
-                "tool_call": True,
-                "limit": {"context": 32000, "output": 8192},
-            },
-        },
-    },
-    "anthropic": {
-        "id": "anthropic",
-        "models": {
-            "claude-sonnet-4": {
-                "id": "claude-sonnet-4",
-                "attachment": True,
-                "tool_call": True,
-                "limit": {"context": 200000, "output": 64000},
-            },
-        },
-    },
-}
-
-
-class TestGetModelCapabilities:
-    """Tests for get_model_capabilities vision detection."""
-
-    def test_vision_from_attachment_flag(self):
-        """Models with attachment=True should report supports_vision=True."""
-        with patch("agent.models_dev.fetch_models_dev", return_value=CAPS_REGISTRY):
-            caps = get_model_capabilities("anthropic", "claude-sonnet-4")
-        assert caps is not None
-        assert caps.supports_vision is True
-
-    def test_vision_from_modalities_input_image(self):
-        """Models with 'image' in modalities.input but attachment=False should
-        still report supports_vision=True (the core fix in this PR)."""
-        with patch("agent.models_dev.fetch_models_dev", return_value=CAPS_REGISTRY):
-            caps = get_model_capabilities("google", "gemma-4-31b-it")
-        assert caps is not None
-        assert caps.supports_vision is True
-
-    def test_no_vision_without_attachment_or_modalities(self):
-        """Models with neither attachment nor image modality should be non-vision."""
-        with patch("agent.models_dev.fetch_models_dev", return_value=CAPS_REGISTRY):
-            caps = get_model_capabilities("google", "gemma-3-1b")
-        assert caps is not None
-        assert caps.supports_vision is False
-
-    def test_modalities_non_dict_handled(self):
-        """Non-dict modalities field should not crash."""
-        registry = {
-            "google": {"id": "google", "models": {
-                "weird-model": {
-                    "id": "weird-model",
-                    "modalities": "text",  # not a dict
-                    "limit": {"context": 200000, "output": 8192},
-                },
-            }},
-        }
-        with patch("agent.models_dev.fetch_models_dev", return_value=registry):
-            caps = get_model_capabilities("gemini", "weird-model")
-        assert caps is not None
-        assert caps.supports_vision is False
-
-    def test_model_not_found_returns_none(self):
-        """Unknown model should return None."""
-        with patch("agent.models_dev.fetch_models_dev", return_value=CAPS_REGISTRY):
-            caps = get_model_capabilities("anthropic", "nonexistent-model")
-        assert caps is None
@@ -1009,4 +1009,65 @@ class TestOpenAIModelExecutionGuidance:
 # =========================================================================


+class TestStripBudgetWarningsFromHistory:
+    def test_strips_json_budget_warning_key(self):
+        import json
+        from run_agent import _strip_budget_warnings_from_history

+        messages = [
+            {"role": "tool", "tool_call_id": "c1", "content": json.dumps({
+                "output": "hello",
+                "exit_code": 0,
+                "_budget_warning": "[BUDGET: Iteration 55/60. 5 iterations left. Start consolidating your work.]",
+            })},
+        ]
+        _strip_budget_warnings_from_history(messages)
+        parsed = json.loads(messages[0]["content"])
+        assert "_budget_warning" not in parsed
+        assert parsed["output"] == "hello"
+        assert parsed["exit_code"] == 0
+
+    def test_strips_text_budget_warning(self):
+        from run_agent import _strip_budget_warnings_from_history
+
+        messages = [
+            {"role": "tool", "tool_call_id": "c1",
+             "content": "some result\n\n[BUDGET WARNING: Iteration 58/60. Only 2 iteration(s) left. Provide your final response NOW. No more tool calls unless absolutely critical.]"},
+        ]
+        _strip_budget_warnings_from_history(messages)
+        assert messages[0]["content"] == "some result"
+
+    def test_leaves_non_tool_messages_unchanged(self):
+        from run_agent import _strip_budget_warnings_from_history
+
+        messages = [
+            {"role": "assistant", "content": "[BUDGET WARNING: Iteration 58/60. Only 2 iteration(s) left. Provide your final response NOW. No more tool calls unless absolutely critical.]"},
+            {"role": "user", "content": "hello"},
+        ]
+        original_contents = [m["content"] for m in messages]
+        _strip_budget_warnings_from_history(messages)
+        assert [m["content"] for m in messages] == original_contents
+
+    def test_handles_empty_and_missing_content(self):
+        from run_agent import _strip_budget_warnings_from_history
+
+        messages = [
+            {"role": "tool", "tool_call_id": "c1", "content": ""},
+            {"role": "tool", "tool_call_id": "c2"},
+        ]
+        _strip_budget_warnings_from_history(messages)
+        assert messages[0]["content"] == ""
+
+    def test_strips_caution_variant(self):
+        import json
+        from run_agent import _strip_budget_warnings_from_history
+
+        messages = [
+            {"role": "tool", "tool_call_id": "c1", "content": json.dumps({
+                "output": "ok",
+                "_budget_warning": "[BUDGET: Iteration 42/60. 18 iterations left. Start consolidating your work.]",
+            })},
+        ]
+        _strip_budget_warnings_from_history(messages)
+        parsed = json.loads(messages[0]["content"])
+        assert "_budget_warning" not in parsed
@@ -1,118 +0,0 @@
-"""Tests for /compress <focus> — guided compression with focus topic.
-
-Inspired by Claude Code's /compact <focus> feature.
-"""
-
-from unittest.mock import MagicMock, patch
-
-from tests.cli.test_cli_init import _make_cli
-
-
-def _make_history() -> list[dict[str, str]]:
-    return [
-        {"role": "user", "content": "one"},
-        {"role": "assistant", "content": "two"},
-        {"role": "user", "content": "three"},
-        {"role": "assistant", "content": "four"},
-    ]
-
-
-def test_focus_topic_extracted_and_passed(capsys):
-    """Focus topic is extracted from the command and passed to _compress_context."""
-    shell = _make_cli()
-    history = _make_history()
-    compressed = [history[0], history[-1]]
-    shell.conversation_history = history
-    shell.agent = MagicMock()
-    shell.agent.compression_enabled = True
-    shell.agent._cached_system_prompt = ""
-    shell.agent._compress_context.return_value = (compressed, "")
-
-    def _estimate(messages):
-        if messages is history:
-            return 100
-        return 50
-
-    with patch("agent.model_metadata.estimate_messages_tokens_rough", side_effect=_estimate):
-        shell._manual_compress("/compress database schema")
-
-    output = capsys.readouterr().out
-    assert 'focus: "database schema"' in output
-
-    # Verify focus_topic was passed through
-    shell.agent._compress_context.assert_called_once()
-    call_kwargs = shell.agent._compress_context.call_args
-    assert call_kwargs.kwargs.get("focus_topic") == "database schema"
-
-
-def test_no_focus_topic_when_bare_command(capsys):
-    """When no focus topic is provided, None is passed."""
-    shell = _make_cli()
-    history = _make_history()
-    shell.conversation_history = history
-    shell.agent = MagicMock()
-    shell.agent.compression_enabled = True
-    shell.agent._cached_system_prompt = ""
-    shell.agent._compress_context.return_value = (list(history), "")
-
-    with patch("agent.model_metadata.estimate_messages_tokens_rough", return_value=100):
-        shell._manual_compress("/compress")
-
-    shell.agent._compress_context.assert_called_once()
-    call_kwargs = shell.agent._compress_context.call_args
-    assert call_kwargs.kwargs.get("focus_topic") is None
-
-
-def test_empty_focus_after_command_treated_as_none(capsys):
-    """Trailing whitespace after /compress does not produce a focus topic."""
-    shell = _make_cli()
-    history = _make_history()
-    shell.conversation_history = history
-    shell.agent = MagicMock()
-    shell.agent.compression_enabled = True
-    shell.agent._cached_system_prompt = ""
-    shell.agent._compress_context.return_value = (list(history), "")
-
-    with patch("agent.model_metadata.estimate_messages_tokens_rough", return_value=100):
-        shell._manual_compress("/compress   ")
-
-    shell.agent._compress_context.assert_called_once()
-    call_kwargs = shell.agent._compress_context.call_args
-    assert call_kwargs.kwargs.get("focus_topic") is None
-
-
-def test_focus_topic_printed_in_compression_banner(capsys):
-    """The focus topic shows in the compression progress banner."""
-    shell = _make_cli()
-    history = _make_history()
-    compressed = [history[0], history[-1]]
-    shell.conversation_history = history
-    shell.agent = MagicMock()
-    shell.agent.compression_enabled = True
-    shell.agent._cached_system_prompt = ""
-    shell.agent._compress_context.return_value = (compressed, "")
-
-    with patch("agent.model_metadata.estimate_messages_tokens_rough", return_value=100):
-        shell._manual_compress("/compress API endpoints")
-
-    output = capsys.readouterr().out
-    assert 'focus: "API endpoints"' in output
-
-
-def test_no_focus_prints_standard_banner(capsys):
-    """Without focus, the standard banner (no focus: line) is printed."""
-    shell = _make_cli()
-    history = _make_history()
-    compressed = [history[0], history[-1]]
-    shell.conversation_history = history
-    shell.agent = MagicMock()
-    shell.agent.compression_enabled = True
-    shell.agent._cached_system_prompt = ""
-    shell.agent._compress_context.return_value = (compressed, "")
-
-    with patch("agent.model_metadata.estimate_messages_tokens_rough", return_value=100):
-        shell._manual_compress("/compress")
-
-    output = capsys.readouterr().out
-    assert "focus:" not in output
-    assert "Compressing" in output
@@ -1,189 +0,0 @@
-"""Tests for stacked tool progress scrollback lines in the CLI TUI.
-
-When tool_progress_mode is "all" or "new", _on_tool_progress should print
-persistent lines to scrollback on tool.completed, restoring the stacked
-tool history that was lost when the TUI switched to a single-line spinner.
-"""
-
-import os
-import sys
-import importlib
-from unittest.mock import MagicMock, patch
-
-sys.path.insert(0, os.path.join(os.path.dirname(__file__), ".."))
-
-# Module-level reference to the cli module (set by _make_cli on first call)
-_cli_mod = None
-
-
-def _make_cli(tool_progress="all"):
-    """Create a HermesCLI instance with minimal mocking."""
-    global _cli_mod
-    _clean_config = {
-        "model": {
-            "default": "anthropic/claude-opus-4.6",
-            "base_url": "https://openrouter.ai/api/v1",
-            "provider": "auto",
-        },
-        "display": {"compact": False, "tool_progress": tool_progress},
-        "agent": {},
-        "terminal": {"env_type": "local"},
-    }
-    clean_env = {"LLM_MODEL": "", "HERMES_MAX_ITERATIONS": ""}
-    prompt_toolkit_stubs = {
-        "prompt_toolkit": MagicMock(),
-        "prompt_toolkit.history": MagicMock(),
-        "prompt_toolkit.styles": MagicMock(),
-        "prompt_toolkit.patch_stdout": MagicMock(),
-        "prompt_toolkit.application": MagicMock(),
-        "prompt_toolkit.layout": MagicMock(),
-        "prompt_toolkit.layout.processors": MagicMock(),
-        "prompt_toolkit.filters": MagicMock(),
-        "prompt_toolkit.layout.dimension": MagicMock(),
-        "prompt_toolkit.layout.menus": MagicMock(),
-        "prompt_toolkit.widgets": MagicMock(),
-        "prompt_toolkit.key_binding": MagicMock(),
-        "prompt_toolkit.completion": MagicMock(),
-        "prompt_toolkit.formatted_text": MagicMock(),
-        "prompt_toolkit.auto_suggest": MagicMock(),
-    }
-    with patch.dict(sys.modules, prompt_toolkit_stubs), \
-         patch.dict("os.environ", clean_env, clear=False):
-        import cli as mod
-        mod = importlib.reload(mod)
-        _cli_mod = mod
-        with patch.object(mod, "get_tool_definitions", return_value=[]), \
-             patch.dict(mod.__dict__, {"CLI_CONFIG": _clean_config}):
-            return mod.HermesCLI()
-
-
-class TestToolProgressScrollback:
-    """Stacked scrollback lines for 'all' and 'new' modes."""
-
-    def test_all_mode_prints_scrollback_on_completed(self):
-        """In 'all' mode, tool.completed prints a stacked line."""
-        cli = _make_cli(tool_progress="all")
-        # Simulate tool.started
-        cli._on_tool_progress("tool.started", "terminal", "git log", {"command": "git log"})
-        # Simulate tool.completed
-        with patch.object(_cli_mod, "_cprint") as mock_print:
-            cli._on_tool_progress("tool.completed", "terminal", None, None, duration=1.5, is_error=False)
-
-        mock_print.assert_called_once()
-        line = mock_print.call_args[0][0]
-        # Should contain tool info (the cute message format has "git log" for terminal)
-        assert "git log" in line or "$" in line
-
-    def test_all_mode_prints_every_call(self):
-        """In 'all' mode, consecutive calls to the same tool each get a line."""
-        cli = _make_cli(tool_progress="all")
-        with patch.object(_cli_mod, "_cprint") as mock_print:
-            # First call
-            cli._on_tool_progress("tool.started", "read_file", "cli.py", {"path": "cli.py"})
-            cli._on_tool_progress("tool.completed", "read_file", None, None, duration=0.1, is_error=False)
-            # Second call (same tool)
-            cli._on_tool_progress("tool.started", "read_file", "run_agent.py", {"path": "run_agent.py"})
-            cli._on_tool_progress("tool.completed", "read_file", None, None, duration=0.2, is_error=False)
-
-        assert mock_print.call_count == 2
-
-    def test_new_mode_skips_consecutive_repeats(self):
-        """In 'new' mode, consecutive calls to the same tool only print once."""
-        cli = _make_cli(tool_progress="new")
-        with patch.object(_cli_mod, "_cprint") as mock_print:
-            cli._on_tool_progress("tool.started", "read_file", "cli.py", {"path": "cli.py"})
-            cli._on_tool_progress("tool.completed", "read_file", None, None, duration=0.1, is_error=False)
-            cli._on_tool_progress("tool.started", "read_file", "run_agent.py", {"path": "run_agent.py"})
-            cli._on_tool_progress("tool.completed", "read_file", None, None, duration=0.2, is_error=False)
-
-        assert mock_print.call_count == 1  # Only the first read_file
-
-    def test_new_mode_prints_when_tool_changes(self):
-        """In 'new' mode, a different tool name triggers a new line."""
-        cli = _make_cli(tool_progress="new")
-        with patch.object(_cli_mod, "_cprint") as mock_print:
-            cli._on_tool_progress("tool.started", "read_file", "cli.py", {"path": "cli.py"})
-            cli._on_tool_progress("tool.completed", "read_file", None, None, duration=0.1, is_error=False)
-            cli._on_tool_progress("tool.started", "search_files", "pattern", {"pattern": "test"})
-            cli._on_tool_progress("tool.completed", "search_files", None, None, duration=0.3, is_error=False)
-            cli._on_tool_progress("tool.started", "read_file", "run_agent.py", {"path": "run_agent.py"})
-            cli._on_tool_progress("tool.completed", "read_file", None, None, duration=0.2, is_error=False)
-
-        # read_file, search_files, read_file (3rd prints because search_files broke the streak)
-        assert mock_print.call_count == 3
-
-    def test_off_mode_no_scrollback(self):
-        """In 'off' mode, no stacked lines are printed."""
-        cli = _make_cli(tool_progress="off")
-        with patch.object(_cli_mod, "_cprint") as mock_print:
-            cli._on_tool_progress("tool.started", "terminal", "ls", {"command": "ls"})
-            cli._on_tool_progress("tool.completed", "terminal", None, None, duration=0.5, is_error=False)
-
-        mock_print.assert_not_called()
-
-    def test_error_suffix_on_failed_tool(self):
-        """When is_error=True, the stacked line includes [error]."""
-        cli = _make_cli(tool_progress="all")
-        cli._on_tool_progress("tool.started", "terminal", "bad cmd", {"command": "bad cmd"})
-        with patch.object(_cli_mod, "_cprint") as mock_print:
-            cli._on_tool_progress("tool.completed", "terminal", None, None, duration=0.5, is_error=True)
-
-        line = mock_print.call_args[0][0]
-        assert "[error]" in line
-
-    def test_spinner_still_updates_on_started(self):
-        """tool.started still updates the spinner text for live display."""
-        cli = _make_cli(tool_progress="all")
-        cli._on_tool_progress("tool.started", "terminal", "git status", {"command": "git status"})
-        assert "git status" in cli._spinner_text
-
-    def test_spinner_timer_clears_on_completed(self):
-        """tool.completed still clears the tool timer."""
-        cli = _make_cli(tool_progress="all")
-        cli._on_tool_progress("tool.started", "terminal", "git status", {"command": "git status"})
-        assert cli._tool_start_time > 0
-        with patch.object(_cli_mod, "_cprint"):
-            cli._on_tool_progress("tool.completed", "terminal", None, None, duration=0.5, is_error=False)
-        assert cli._tool_start_time == 0.0
-
-    def test_concurrent_tools_produce_stacked_lines(self):
-        """Multiple tool.started followed by multiple tool.completed all produce lines."""
-        cli = _make_cli(tool_progress="all")
-        with patch.object(_cli_mod, "_cprint") as mock_print:
-            # All start first (concurrent pattern)
-            cli._on_tool_progress("tool.started", "web_search", "query 1", {"query": "test 1"})
-            cli._on_tool_progress("tool.started", "web_search", "query 2", {"query": "test 2"})
-            # All complete
-            cli._on_tool_progress("tool.completed", "web_search", None, None, duration=1.0, is_error=False)
-            cli._on_tool_progress("tool.completed", "web_search", None, None, duration=1.5, is_error=False)
-
-        assert mock_print.call_count == 2
-
-    def test_verbose_mode_no_duplicate_scrollback(self):
-        """In 'verbose' mode, scrollback lines are NOT printed (run_agent handles verbose output)."""
-        cli = _make_cli(tool_progress="verbose")
-        with patch.object(_cli_mod, "_cprint") as mock_print:
-            cli._on_tool_progress("tool.started", "terminal", "ls", {"command": "ls"})
-            cli._on_tool_progress("tool.completed", "terminal", None, None, duration=0.5, is_error=False)
-
-        mock_print.assert_not_called()
-
-    def test_pending_info_stores_on_started(self):
-        """tool.started stores args for later use by tool.completed."""
-        cli = _make_cli(tool_progress="all")
-        cli._on_tool_progress("tool.started", "terminal", "ls", {"command": "ls"})
-        assert "terminal" in cli._pending_tool_info
-        assert len(cli._pending_tool_info["terminal"]) == 1
-        assert cli._pending_tool_info["terminal"][0] == {"command": "ls"}
-
-    def test_pending_info_consumed_on_completed(self):
-        """tool.completed consumes stored args (FIFO for concurrent)."""
-        cli = _make_cli(tool_progress="all")
-        cli._on_tool_progress("tool.started", "terminal", "ls", {"command": "ls"})
-        cli._on_tool_progress("tool.started", "terminal", "pwd", {"command": "pwd"})
-        assert len(cli._pending_tool_info["terminal"]) == 2
-        with patch.object(_cli_mod, "_cprint"):
-            cli._on_tool_progress("tool.completed", "terminal", None, None, duration=0.1, is_error=False)
-        # First entry consumed, second remains
-        assert len(cli._pending_tool_info.get("terminal", [])) == 1
-        assert cli._pending_tool_info["terminal"][0] == {"command": "pwd"}
@@ -211,8 +211,7 @@ def make_adapter(platform: Platform, runner=None):
    config = PlatformConfig(enabled=True, token="e2e-test-token")

    if platform == Platform.DISCORD:
-        from gateway.platforms.helpers import ThreadParticipationTracker
-        with patch.object(ThreadParticipationTracker, "_load", return_value=set()):
+        with patch.object(DiscordAdapter, "_load_participated_threads", return_value=set()):
            adapter = DiscordAdapter(config)
        platform_key = Platform.DISCORD
    elif platform == Platform.SLACK:
@@ -409,50 +409,11 @@ class TestChatCompletionsEndpoint:
                )
                assert resp.status == 200
                assert "text/event-stream" in resp.headers.get("Content-Type", "")
-                assert resp.headers.get("X-Accel-Buffering") == "no"
                body = await resp.text()
                assert "data: " in body
                assert "[DONE]" in body
                assert "Hello!" in body

-    @pytest.mark.asyncio
-    async def test_stream_sends_keepalive_during_quiet_tool_gap(self, adapter):
-        """Idle SSE streams should send keepalive comments while tools run silently."""
-        import asyncio
-        import gateway.platforms.api_server as api_server_mod
-
-        app = _create_app(adapter)
-        async with TestClient(TestServer(app)) as cli:
-            async def _mock_run_agent(**kwargs):
-                cb = kwargs.get("stream_delta_callback")
-                if cb:
-                    cb("Working")
-                    await asyncio.sleep(0.65)
-                    cb("...done")
-                return (
-                    {"final_response": "Working...done", "messages": [], "api_calls": 1},
-                    {"input_tokens": 10, "output_tokens": 5, "total_tokens": 15},
-                )
-
-            with (
-                patch.object(api_server_mod, "CHAT_COMPLETIONS_SSE_KEEPALIVE_SECONDS", 0.01),
-                patch.object(adapter, "_run_agent", side_effect=_mock_run_agent),
-            ):
-                resp = await cli.post(
-                    "/v1/chat/completions",
-                    json={
-                        "model": "test",
-                        "messages": [{"role": "user", "content": "do the thing"}],
-                        "stream": True,
-                    },
-                )
-                assert resp.status == 200
-                body = await resp.text()
-                assert ": keepalive" in body
-                assert "Working" in body
-                assert "...done" in body
-                assert "[DONE]" in body
-
    @pytest.mark.asyncio
    async def test_stream_survives_tool_call_none_sentinel(self, adapter):
        """stream_delta_callback(None) mid-stream (tool calls) must NOT kill the SSE stream.
@@ -345,11 +345,6 @@ class TestBlockingApprovalE2E:

    def setup_method(self):
        _clear_approval_state()
-        os.environ.pop("HERMES_YOLO_MODE", None)
-        os.environ.pop("HERMES_INTERACTIVE", None)
-        os.environ.pop("HERMES_GATEWAY_SESSION", None)
-        os.environ.pop("HERMES_EXEC_ASK", None)
-        os.environ.pop("HERMES_SESSION_KEY", None)

    def test_blocking_approval_approve_once(self):
        """check_all_command_guards blocks until resolve_gateway_approval is called."""
@@ -369,7 +364,6 @@ class TestBlockingApprovalE2E:
            from tools.approval import reset_current_session_key, set_current_session_key

            token = set_current_session_key(session_key)
-            os.environ["HERMES_GATEWAY_SESSION"] = "1"
            os.environ["HERMES_EXEC_ASK"] = "1"
            os.environ["HERMES_SESSION_KEY"] = session_key
            try:
@@ -377,7 +371,6 @@ class TestBlockingApprovalE2E:
                    "rm -rf /important", "local"
                )
            finally:
-                os.environ.pop("HERMES_GATEWAY_SESSION", None)
                os.environ.pop("HERMES_EXEC_ASK", None)
                os.environ.pop("HERMES_SESSION_KEY", None)
                reset_current_session_key(token)
@@ -417,7 +410,6 @@ class TestBlockingApprovalE2E:
            from tools.approval import reset_current_session_key, set_current_session_key

            token = set_current_session_key(session_key)
-            os.environ["HERMES_GATEWAY_SESSION"] = "1"
            os.environ["HERMES_EXEC_ASK"] = "1"
            os.environ["HERMES_SESSION_KEY"] = session_key
            try:
@@ -425,7 +417,6 @@ class TestBlockingApprovalE2E:
                    "rm -rf /important", "local"
                )
            finally:
-                os.environ.pop("HERMES_GATEWAY_SESSION", None)
                os.environ.pop("HERMES_EXEC_ASK", None)
                os.environ.pop("HERMES_SESSION_KEY", None)
                reset_current_session_key(token)
@@ -460,7 +451,6 @@ class TestBlockingApprovalE2E:
            from tools.approval import reset_current_session_key, set_current_session_key

            token = set_current_session_key(session_key)
-            os.environ["HERMES_GATEWAY_SESSION"] = "1"
            os.environ["HERMES_EXEC_ASK"] = "1"
            os.environ["HERMES_SESSION_KEY"] = session_key
            try:
@@ -470,7 +460,6 @@ class TestBlockingApprovalE2E:
                        "rm -rf /important", "local"
                    )
            finally:
-                os.environ.pop("HERMES_GATEWAY_SESSION", None)
                os.environ.pop("HERMES_EXEC_ASK", None)
                os.environ.pop("HERMES_SESSION_KEY", None)
                reset_current_session_key(token)
@@ -502,13 +491,11 @@ class TestBlockingApprovalE2E:
                from tools.approval import reset_current_session_key, set_current_session_key

                token = set_current_session_key(session_key)
-                os.environ["HERMES_GATEWAY_SESSION"] = "1"
                os.environ["HERMES_EXEC_ASK"] = "1"
                os.environ["HERMES_SESSION_KEY"] = session_key
                try:
                    results[idx] = check_all_command_guards(cmd, "local")
                finally:
-                    os.environ.pop("HERMES_GATEWAY_SESSION", None)
                    os.environ.pop("HERMES_EXEC_ASK", None)
                    os.environ.pop("HERMES_SESSION_KEY", None)
                    reset_current_session_key(token)
@@ -559,13 +546,11 @@ class TestBlockingApprovalE2E:
                from tools.approval import reset_current_session_key, set_current_session_key

                token = set_current_session_key(session_key)
-                os.environ["HERMES_GATEWAY_SESSION"] = "1"
                os.environ["HERMES_EXEC_ASK"] = "1"
                os.environ["HERMES_SESSION_KEY"] = session_key
                try:
                    results[idx] = check_all_command_guards(cmd, "local")
                finally:
-                    os.environ.pop("HERMES_GATEWAY_SESSION", None)
                    os.environ.pop("HERMES_EXEC_ASK", None)
                    os.environ.pop("HERMES_SESSION_KEY", None)
                    reset_current_session_key(token)
@@ -1,118 +0,0 @@
-"""Tests for gateway /compress <focus> — focus topic on the gateway side."""
-
-from datetime import datetime
-from unittest.mock import MagicMock, patch
-
-import pytest
-
-from gateway.config import GatewayConfig, Platform, PlatformConfig
-from gateway.platforms.base import MessageEvent
-from gateway.session import SessionEntry, SessionSource, build_session_key
-
-
-def _make_source() -> SessionSource:
-    return SessionSource(
-        platform=Platform.TELEGRAM,
-        user_id="u1",
-        chat_id="c1",
-        user_name="tester",
-        chat_type="dm",
-    )
-
-
-def _make_event(text: str = "/compress") -> MessageEvent:
-    return MessageEvent(text=text, source=_make_source(), message_id="m1")
-
-
-def _make_history() -> list[dict[str, str]]:
-    return [
-        {"role": "user", "content": "one"},
-        {"role": "assistant", "content": "two"},
-        {"role": "user", "content": "three"},
-        {"role": "assistant", "content": "four"},
-    ]
-
-
-def _make_runner(history: list[dict[str, str]]):
-    from gateway.run import GatewayRunner
-
-    runner = object.__new__(GatewayRunner)
-    runner.config = GatewayConfig(
-        platforms={Platform.TELEGRAM: PlatformConfig(enabled=True, token="***")}
-    )
-    session_entry = SessionEntry(
-        session_key=build_session_key(_make_source()),
-        session_id="sess-1",
-        created_at=datetime.now(),
-        updated_at=datetime.now(),
-        platform=Platform.TELEGRAM,
-        chat_type="dm",
-    )
-    runner.session_store = MagicMock()
-    runner.session_store.get_or_create_session.return_value = session_entry
-    runner.session_store.load_transcript.return_value = history
-    runner.session_store.rewrite_transcript = MagicMock()
-    runner.session_store.update_session = MagicMock()
-    runner.session_store._save = MagicMock()
-    return runner
-
-
-@pytest.mark.asyncio
-async def test_compress_focus_topic_passed_to_agent():
-    """Focus topic from /compress <focus> is passed through to _compress_context."""
-    history = _make_history()
-    compressed = [history[0], history[-1]]
-    runner = _make_runner(history)
-    agent_instance = MagicMock()
-    agent_instance.context_compressor.protect_first_n = 0
-    agent_instance.context_compressor._align_boundary_forward.return_value = 0
-    agent_instance.context_compressor._find_tail_cut_by_tokens.return_value = 2
-    agent_instance.session_id = "sess-1"
-    agent_instance._compress_context.return_value = (compressed, "")
-
-    def _estimate(messages):
-        return 100
-
-    with (
-        patch("gateway.run._resolve_runtime_agent_kwargs", return_value={"api_key": "***"}),
-        patch("gateway.run._resolve_gateway_model", return_value="test-model"),
-        patch("run_agent.AIAgent", return_value=agent_instance),
-        patch("agent.model_metadata.estimate_messages_tokens_rough", side_effect=_estimate),
-    ):
-        result = await runner._handle_compress_command(_make_event("/compress database schema"))
-
-    # Verify focus_topic was passed
-    agent_instance._compress_context.assert_called_once()
-    call_kwargs = agent_instance._compress_context.call_args
-    assert call_kwargs.kwargs.get("focus_topic") == "database schema"
-
-    # Verify focus is mentioned in response
-    assert 'Focus: "database schema"' in result
-
-
-@pytest.mark.asyncio
-async def test_compress_no_focus_passes_none():
-    """Bare /compress passes focus_topic=None."""
-    history = _make_history()
-    runner = _make_runner(history)
-    agent_instance = MagicMock()
-    agent_instance.context_compressor.protect_first_n = 0
-    agent_instance.context_compressor._align_boundary_forward.return_value = 0
-    agent_instance.context_compressor._find_tail_cut_by_tokens.return_value = 2
-    agent_instance.session_id = "sess-1"
-    agent_instance._compress_context.return_value = (list(history), "")
-
-    with (
-        patch("gateway.run._resolve_runtime_agent_kwargs", return_value={"api_key": "***"}),
-        patch("gateway.run._resolve_gateway_model", return_value="test-model"),
-        patch("run_agent.AIAgent", return_value=agent_instance),
-        patch("agent.model_metadata.estimate_messages_tokens_rough", return_value=100),
-    ):
-        result = await runner._handle_compress_command(_make_event("/compress"))
-
-    agent_instance._compress_context.assert_called_once()
-    call_kwargs = agent_instance._compress_context.call_args
-    assert call_kwargs.kwargs.get("focus_topic") is None
-
-    # No focus line in response
-    assert "Focus:" not in result
--- a/Show More
+++ b/Show More