fix(moonshot): strip $ref siblings and collapse tuple items in tool schemas

Port from anomalyco/opencode#24730: Moonshot's JSON Schema validator rejects
two shapes that the rest of the JSON Schema ecosystem accepts:

1. $ref nodes with sibling keywords. Moonshot expands the reference before
   validation and then rejects the node if keys like `description`, `type`,
   or `default` appear alongside $ref. MCP-sourced tool schemas commonly
   put a `description` on $ref-typed properties so the model sees the
   field hint — which worked on every provider except Moonshot.

2. Tuple-style `items` arrays (positional element schemas). Moonshot's
   engine requires ONE schema applied to every array element. Common in
   tool schemas generated from Go/Protobuf that model fixed-length arrays
   as `[{type:number}, {type:number}]`.

Repairs applied in `agent/moonshot_schema.py`:

- Rule 3: when a node has `$ref`, return `{"$ref": <value>}` only
  (strip every sibling). The referenced definition still carries its own
  description on the target node, which Moonshot accepts.
- Rule 4: when `items` is a list, collapse to the first element schema
  (falling back to `{}` which is then filled by the generic missing-type
  rule). Preserves `minItems` / `maxItems` / other siblings.

Tests: 10 new cases across TestRefSiblingStripping + TestTupleItems,
plus the existing TestMissingTypeFilled::test_ref_node_is_not_given_synthetic_type
still passes (it asserted plain $ref passes through; now it passes through
as exactly `{"$ref": "..."}` which is strictly compatible).

All 35 tests in test_moonshot_schema.py pass.

.dockerignore

@@ -9,12 +9,6 @@ node_modules
 .venv
 **/.venv
 
-# Built artifacts that are regenerated inside the image.  Excluded so local
-# rebuilds on the developer's machine don't invalidate the npm-install layer
-# that now depends on the full ui-tui/packages/hermes-ink/ tree being present.
-ui-tui/dist/
-ui-tui/packages/hermes-ink/dist/
-
 # CI/CD
 .github
 

.github/workflows/deploy-site.yml

@@ -76,16 +76,6 @@ jobs:
         run: |
           mkdir -p _site/docs
           cp -r website/build/* _site/docs/
-          # llms.txt / llms-full.txt are also published at the site root
-          # (https://hermes-agent.nousresearch.com/llms.txt) because some
-          # agents and IDE plugins probe the classic root-level path rather
-          # than /docs/llms.txt. Same file, two URLs, one source of truth.
-          if [ -f website/build/llms.txt ]; then
-            cp website/build/llms.txt _site/llms.txt
-          fi
-          if [ -f website/build/llms-full.txt ]; then
-            cp website/build/llms-full.txt _site/llms-full.txt
-          fi
 
       - name: Upload artifact
         uses: actions/upload-pages-artifact@56afc609e74202658d3ffba0e8f6dda462b719fa  # v3

Dockerfile

@@ -28,26 +28,10 @@ WORKDIR /opt/hermes
 # ---------- Layer-cached dependency install ----------
 # Copy only package manifests first so npm install + Playwright are cached
 # unless the lockfiles themselves change.
-#
-# ui-tui/packages/hermes-ink/ is copied IN FULL (not just its manifests)
-# because it is referenced as a `file:` workspace dependency from
-# ui-tui/package.json.  Copying the tree up front lets npm resolve the
-# workspace to real content instead of stopping at a bare package.json.
 COPY package.json package-lock.json ./
 COPY web/package.json web/package-lock.json web/
 COPY ui-tui/package.json ui-tui/package-lock.json ui-tui/
-COPY ui-tui/packages/hermes-ink/ ui-tui/packages/hermes-ink/
-
-# `npm_config_install_links=false` forces npm to install `file:` deps as
-# symlinks (the npm 10+ default) even on Debian's older bundled npm 9.x,
-# which defaults to `install-links=true` and installs file deps as *copies*.
-# The host-side package-lock.json is generated with a newer npm that uses
-# symlinks, so an install-as-copy produces a hidden node_modules/.package-lock.json
-# that permanently disagrees with the root lock on the @hermes/ink entry.
-# That disagreement trips the TUI launcher's `_tui_need_npm_install()`
-# check on every startup and triggers a runtime `npm install` that then
-# fails with EACCES (node_modules/ is root-owned from build time).
-ENV npm_config_install_links=false
+COPY ui-tui/packages/hermes-ink/package.json ui-tui/packages/hermes-ink/package-lock.json ui-tui/packages/hermes-ink/
 
 RUN npm install --prefer-offline --no-audit && \
     npx playwright install --with-deps chromium --only-shell && \
@@ -61,7 +45,13 @@ COPY --chown=hermes:hermes . .
 
 # Build browser dashboard and terminal UI assets.
 RUN cd web && npm run build && \
-    cd ../ui-tui && npm run build
+    cd ../ui-tui && npm run build && \
+    rm -rf node_modules/@hermes/ink && \
+    rm -rf packages/hermes-ink/node_modules && \
+    cp -R packages/hermes-ink node_modules/@hermes/ink && \
+    npm install --omit=dev --prefer-offline --no-audit --prefix node_modules/@hermes/ink && \
+    rm -rf node_modules/@hermes/ink/node_modules/react && \
+    node --input-type=module -e "await import('@hermes/ink')"
 
 # ---------- Permissions ----------
 # Make install dir world-readable so any HERMES_UID can read it at runtime.

acp_adapter/server.py

@@ -164,8 +164,6 @@ class HermesACPAgent(acp.Agent):
         "context": "Show conversation context info",
         "reset": "Clear conversation history",
         "compact": "Compress conversation context",
-        "steer": "Inject guidance into the currently running agent turn",
-        "queue": "Queue a prompt to run after the current turn finishes",
         "version": "Show Hermes version",
     }
 
@@ -195,16 +193,6 @@ class HermesACPAgent(acp.Agent):
             "name": "compact",
             "description": "Compress conversation context",
         },
-        {
-            "name": "steer",
-            "description": "Inject guidance into the currently running agent turn",
-            "input_hint": "guidance for the active turn",
-        },
-        {
-            "name": "queue",
-            "description": "Queue a prompt to run after the current turn finishes",
-            "input_hint": "prompt to run next",
-        },
         {
             "name": "version",
             "description": "Show Hermes version",
@@ -569,9 +557,6 @@ class HermesACPAgent(acp.Agent):
     async def cancel(self, session_id: str, **kwargs: Any) -> None:
         state = self.session_manager.get_session(session_id)
         if state and state.cancel_event:
-            with state.runtime_lock:
-                if state.is_running and state.current_prompt_text:
-                    state.interrupted_prompt_text = state.current_prompt_text
             state.cancel_event.set()
             try:
                 if getattr(state, "agent", None) and hasattr(state.agent, "interrupt"):
@@ -669,39 +654,6 @@ class HermesACPAgent(acp.Agent):
         if not has_content:
             return PromptResponse(stop_reason="end_turn")
 
-        # /steer on an idle session has no in-flight tool call to inject into.
-        # Rewrite it so the payload runs as a normal user prompt, matching the
-        # gateway's behavior (gateway/run.py ~L4898). Two sub-cases:
-        #   1. Zed-interrupt salvage — a prior prompt was cancelled by the
-        #      client right before /steer arrived; replay it with the steer
-        #      text attached as explicit correction/guidance so the user's
-        #      in-flight work isn't lost.
-        #   2. Plain idle — no prior work to salvage; just run the steer
-        #      payload as a regular prompt. Without this, _cmd_steer would
-        #      silently append to state.queued_prompts and respond with
-        #      "No active turn — queued for the next turn", which looks like
-        #      /queue even though the user never typed /queue.
-        if isinstance(user_content, str) and user_text.startswith("/steer"):
-            steer_text = user_text.split(maxsplit=1)[1].strip() if len(user_text.split(maxsplit=1)) > 1 else ""
-            interrupted_prompt = ""
-            rewrite_idle = False
-            with state.runtime_lock:
-                if not state.is_running and steer_text:
-                    if state.interrupted_prompt_text:
-                        interrupted_prompt = state.interrupted_prompt_text
-                        state.interrupted_prompt_text = ""
-                    else:
-                        rewrite_idle = True
-            if interrupted_prompt:
-                user_text = (
-                    f"{interrupted_prompt}\n\n"
-                    f"User correction/guidance after interrupt: {steer_text}"
-                )
-                user_content = user_text
-            elif rewrite_idle:
-                user_text = steer_text
-                user_content = steer_text
-
         # Intercept slash commands — handle locally without calling the LLM.
         # Slash commands are text-only; if the client included images/resources,
         # send the whole multimodal prompt to the agent instead of treating it as
@@ -714,24 +666,6 @@ class HermesACPAgent(acp.Agent):
                     await self._conn.session_update(session_id, update)
                 return PromptResponse(stop_reason="end_turn")
 
-        # If Zed sends another regular prompt while the same ACP session is
-        # still running, queue it instead of racing two AIAgent loops against
-        # the same state.history. /steer and /queue are handled above and can
-        # land immediately.
-        with state.runtime_lock:
-            if state.is_running:
-                queued_text = user_text or "[Image attachment]"
-                state.queued_prompts.append(queued_text)
-                depth = len(state.queued_prompts)
-                if self._conn:
-                    update = acp.update_agent_message_text(
-                        f"Queued for the next turn. ({depth} queued)"
-                    )
-                    await self._conn.session_update(session_id, update)
-                return PromptResponse(stop_reason="end_turn")
-            state.is_running = True
-            state.current_prompt_text = user_text or "[Image attachment]"
-
         logger.info("Prompt on session %s: %s", session_id, user_text[:100])
 
         conn = self._conn
@@ -843,9 +777,6 @@ class HermesACPAgent(acp.Agent):
             result = await loop.run_in_executor(_executor, ctx.run, _run_agent)
         except Exception:
             logger.exception("Executor error for session %s", session_id)
-            with state.runtime_lock:
-                state.is_running = False
-                state.current_prompt_text = ""
             return PromptResponse(stop_reason="end_turn")
 
         if result.get("messages"):
@@ -871,28 +802,6 @@ class HermesACPAgent(acp.Agent):
             update = acp.update_agent_message_text(final_response)
             await conn.session_update(session_id, update)
 
-        # Mark this turn idle before draining queued work so recursive prompt()
-        # calls can acquire the session. Queued turns are intentionally run as
-        # normal follow-up user prompts, preserving role alternation and history.
-        with state.runtime_lock:
-            state.is_running = False
-            state.current_prompt_text = ""
-
-        while True:
-            with state.runtime_lock:
-                if not state.queued_prompts:
-                    break
-                next_prompt = state.queued_prompts.pop(0)
-            if conn:
-                await conn.session_update(
-                    session_id,
-                    acp.update_user_message_text(next_prompt),
-                )
-            await self.prompt(
-                prompt=[TextContentBlock(type="text", text=next_prompt)],
-                session_id=session_id,
-            )
-
         usage = None
         if any(result.get(key) is not None for key in ("prompt_tokens", "completion_tokens", "total_tokens")):
             usage = Usage(
@@ -970,8 +879,6 @@ class HermesACPAgent(acp.Agent):
             "context": self._cmd_context,
             "reset": self._cmd_reset,
             "compact": self._cmd_compact,
-            "steer": self._cmd_steer,
-            "queue": self._cmd_queue,
             "version": self._cmd_version,
         }.get(cmd)
 
@@ -1068,16 +975,10 @@ class HermesACPAgent(acp.Agent):
             if not hasattr(agent, "_compress_context"):
                 return "Context compression not available for this agent."
 
-            from agent.model_metadata import estimate_request_tokens_rough
+            from agent.model_metadata import estimate_messages_tokens_rough
 
             original_count = len(state.history)
-            # Include system prompt + tool schemas so the figure reflects real
-            # request pressure, not a transcript-only underestimate (#6217).
-            _sys_prompt = getattr(agent, "_cached_system_prompt", "") or ""
-            _tools = getattr(agent, "tools", None) or None
-            approx_tokens = estimate_request_tokens_rough(
-                state.history, system_prompt=_sys_prompt, tools=_tools
-            )
+            approx_tokens = estimate_messages_tokens_rough(state.history)
             original_session_db = getattr(agent, "_session_db", None)
 
             try:
@@ -1097,13 +998,7 @@ class HermesACPAgent(acp.Agent):
             self.session_manager.save_session(state.session_id)
 
             new_count = len(state.history)
-            _sys_prompt_after = getattr(agent, "_cached_system_prompt", "") or _sys_prompt
-            _tools_after = getattr(agent, "tools", None) or _tools
-            new_tokens = estimate_request_tokens_rough(
-                state.history,
-                system_prompt=_sys_prompt_after,
-                tools=_tools_after,
-            )
+            new_tokens = estimate_messages_tokens_rough(state.history)
             return (
                 f"Context compressed: {original_count} -> {new_count} messages\n"
                 f"~{approx_tokens:,} -> ~{new_tokens:,} tokens"
@@ -1111,34 +1006,6 @@ class HermesACPAgent(acp.Agent):
         except Exception as e:
             return f"Compression failed: {e}"
 
-    def _cmd_steer(self, args: str, state: SessionState) -> str:
-        steer_text = args.strip()
-        if not steer_text:
-            return "Usage: /steer <guidance>"
-
-        if state.is_running and hasattr(state.agent, "steer"):
-            try:
-                if state.agent.steer(steer_text):
-                    preview = steer_text[:80] + ("..." if len(steer_text) > 80 else "")
-                    return f"⏩ Steer queued for the active turn: {preview}"
-            except Exception as exc:
-                logger.warning("ACP steer failed for session %s: %s", state.session_id, exc)
-                return f"⚠️ Steer failed: {exc}"
-
-        with state.runtime_lock:
-            state.queued_prompts.append(steer_text)
-            depth = len(state.queued_prompts)
-        return f"No active turn — queued for the next turn. ({depth} queued)"
-
-    def _cmd_queue(self, args: str, state: SessionState) -> str:
-        queued_text = args.strip()
-        if not queued_text:
-            return "Usage: /queue <prompt>"
-        with state.runtime_lock:
-            state.queued_prompts.append(queued_text)
-            depth = len(state.queued_prompts)
-        return f"Queued for the next turn. ({depth} queued)"
-
     def _cmd_version(self, args: str, state: SessionState) -> str:
         return f"Hermes Agent v{HERMES_VERSION}"
 

acp_adapter/session.py

@@ -26,33 +26,6 @@ from typing import Any, Dict, List, Optional
 logger = logging.getLogger(__name__)
 
 
-def _win_path_to_wsl(path: str) -> str | None:
-    """Convert a Windows drive path to its WSL /mnt/<drive>/... equivalent."""
-    match = re.match(r"^([A-Za-z]):[\\/](.*)$", path)
-    if not match:
-        return None
-    drive = match.group(1).lower()
-    tail = match.group(2).replace("\\", "/")
-    return f"/mnt/{drive}/{tail}"
-
-
-def _translate_acp_cwd(cwd: str) -> str:
-    """Translate Windows ACP cwd values when Hermes itself is running in WSL.
-
-    Windows ACP clients can launch ``hermes acp`` inside WSL while still sending
-    editor workspaces as Windows drive paths such as ``E:\\Projects``. Store
-    and execute against the WSL mount path so agents, tools, and persisted ACP
-    sessions all agree on the usable workspace. Native Linux/macOS keeps the
-    original cwd unchanged.
-    """
-    from hermes_constants import is_wsl
-
-    if not is_wsl():
-        return cwd
-    translated = _win_path_to_wsl(str(cwd))
-    return translated if translated is not None else cwd
-
-
 def _normalize_cwd_for_compare(cwd: str | None) -> str:
     raw = str(cwd or ".").strip()
     if not raw:
@@ -61,9 +34,11 @@ def _normalize_cwd_for_compare(cwd: str | None) -> str:
 
     # Normalize Windows drive paths into the equivalent WSL mount form so
     # ACP history filters match the same workspace across Windows and WSL.
-    translated = _win_path_to_wsl(expanded)
-    if translated is not None:
-        expanded = translated
+    match = re.match(r"^([A-Za-z]):[\\/](.*)$", expanded)
+    if match:
+        drive = match.group(1).lower()
+        tail = match.group(2).replace("\\", "/")
+        expanded = f"/mnt/{drive}/{tail}"
     elif re.match(r"^/mnt/[A-Za-z]/", expanded):
         expanded = f"/mnt/{expanded[5].lower()}/{expanded[7:]}"
 
@@ -121,18 +96,12 @@ def _acp_stderr_print(*args, **kwargs) -> None:
 
 
 def _register_task_cwd(task_id: str, cwd: str) -> None:
-    """Bind a task/session id to the editor's working directory for tools.
-
-    Zed can launch Hermes from a Windows workspace while the ACP process runs
-    inside WSL. In that case ACP sends cwd as e.g. ``E:\\Projects\\POTI``;
-    local tools need the WSL mount equivalent or subprocess creation fails
-    before the command can run.
-    """
+    """Bind a task/session id to the editor's working directory for tools."""
     if not task_id:
         return
     try:
         from tools.terminal_tool import register_task_env_overrides
-        register_task_env_overrides(task_id, {"cwd": _translate_acp_cwd(cwd)})
+        register_task_env_overrides(task_id, {"cwd": cwd})
     except Exception:
         logger.debug("Failed to register ACP task cwd override", exc_info=True)
 
@@ -176,11 +145,6 @@ class SessionState:
     model: str = ""
     history: List[Dict[str, Any]] = field(default_factory=list)
     cancel_event: Any = None  # threading.Event
-    is_running: bool = False
-    queued_prompts: List[str] = field(default_factory=list)
-    runtime_lock: Any = field(default_factory=Lock)
-    current_prompt_text: str = ""
-    interrupted_prompt_text: str = ""
 
 
 class SessionManager:
@@ -211,7 +175,6 @@ class SessionManager:
         """Create a new session with a unique ID and a fresh AIAgent."""
         import threading
 
-        cwd = _translate_acp_cwd(cwd)
         session_id = str(uuid.uuid4())
         agent = self._make_agent(session_id=session_id, cwd=cwd)
         state = SessionState(
@@ -254,7 +217,6 @@ class SessionManager:
         """Deep-copy a session's history into a new session."""
         import threading
 
-        cwd = _translate_acp_cwd(cwd)
         original = self.get_session(session_id)  # checks DB too
         if original is None:
             return None
@@ -356,7 +318,6 @@ class SessionManager:
 
     def update_cwd(self, session_id: str, cwd: str) -> Optional[SessionState]:
         """Update the working directory for a session and its tool overrides."""
-        cwd = _translate_acp_cwd(cwd)
         state = self.get_session(session_id)  # checks DB too
         if state is None:
             return None

agent/auxiliary_client.py

@@ -1977,12 +1977,6 @@ def resolve_provider_client(
         (client, resolved_model) or (None, None) if auth is unavailable.
     """
     _validate_proxy_env_urls()
-    # Preserve the original provider name before alias normalization so a
-    # user-declared ``custom_providers`` entry whose name coincidentally
-    # matches a built-in alias (e.g. user names their custom provider "kimi"
-    # which aliases to "kimi-coding") is still reachable via the named-custom
-    # branch below.
-    original_provider = (provider or "").strip().lower()
     # Normalise aliases
     provider = _normalize_aux_provider(provider)
 
@@ -2169,18 +2163,7 @@ def resolve_provider_client(
     # ── Named custom providers (config.yaml providers dict / custom_providers list) ───
     try:
         from hermes_cli.runtime_provider import _get_named_custom_provider
-        # When the raw requested name is an alias (``kimi`` → ``kimi-coding``)
-        # and the user defined a ``custom_providers`` entry under that alias
-        # name, the custom entry is the intended target — the built-in alias
-        # rewriting would otherwise hijack the request.  Only preferred when
-        # the raw name is an alias (not a canonical provider name) so custom
-        # entries that coincidentally match a canonical provider (e.g. ``nous``)
-        # still defer to the built-in per `_get_named_custom_provider`'s guard.
-        custom_entry = None
-        if original_provider and original_provider != provider:
-            custom_entry = _get_named_custom_provider(original_provider)
-        if custom_entry is None:
-            custom_entry = _get_named_custom_provider(provider)
+        custom_entry = _get_named_custom_provider(provider)
         if custom_entry:
             custom_base = custom_entry.get("base_url", "").strip()
             custom_key = custom_entry.get("api_key", "").strip()
@@ -2290,12 +2273,6 @@ def resolve_provider_client(
 
         creds = resolve_api_key_provider_credentials(provider)
         api_key = str(creds.get("api_key", "")).strip()
-        # Honour an explicit api_key override (e.g. from a fallback_model entry
-        # or a custom_providers entry) so callers that pass an explicit
-        # credential can authenticate against endpoints where no built-in
-        # credential is registered for this provider alias.
-        if explicit_api_key:
-            api_key = explicit_api_key.strip() or api_key
         if not api_key:
             tried_sources = list(pconfig.api_key_env_vars)
             if provider == "copilot":
@@ -2307,11 +2284,6 @@ def resolve_provider_client(
 
         raw_base_url = str(creds.get("base_url", "")).strip().rstrip("/") or pconfig.inference_base_url
         base_url = _to_openai_base_url(raw_base_url)
-        # Honour an explicit base_url override from the caller — used when a
-        # fallback_model entry (or custom_providers lookup) routes through a
-        # built-in provider name but targets a user-specified endpoint.
-        if explicit_base_url:
-            base_url = _to_openai_base_url(explicit_base_url.strip().rstrip("/"))
 
         default_model = _API_KEY_PROVIDER_AUX_MODELS.get(provider, "")
         final_model = _normalize_resolved_model(model or default_model, provider)

agent/context_compressor.py

@@ -538,7 +538,7 @@ class ContextCompressor(ContextEngine):
             # Token-budget approach: walk backward accumulating tokens
             accumulated = 0
             boundary = len(result)
-            min_protect = min(protect_tail_count, len(result))
+            min_protect = min(protect_tail_count, len(result) - 1)
             for i in range(len(result) - 1, -1, -1):
                 msg = result[i]
                 raw_content = msg.get("content") or ""
@@ -992,8 +992,8 @@ The user has requested that this compaction PRIORITISE preserving all informatio
     def _get_tool_call_id(tc) -> str:
         """Extract the call ID from a tool_call entry (dict or SimpleNamespace)."""
         if isinstance(tc, dict):
-            return tc.get("call_id", "") or tc.get("id", "") or ""
-        return getattr(tc, "call_id", "") or getattr(tc, "id", "") or ""
+            return tc.get("id", "")
+        return getattr(tc, "id", "") or ""
 
     def _sanitize_tool_pairs(self, messages: List[Dict[str, Any]]) -> List[Dict[str, Any]]:
         """Fix orphaned tool_call / tool_result pairs after compression.

agent/curator.py

@@ -55,7 +55,6 @@ def _default_state() -> Dict[str, Any]:
         "last_run_at": None,
         "last_run_duration_seconds": None,
         "last_run_summary": None,
-        "last_report_path": None,
         "paused": False,
         "run_count": 0,
     }
@@ -767,39 +766,6 @@ def _write_run_report(
     consolidated = classification["consolidated"]
     pruned = classification["pruned"]
 
-    # Rewrite cron job skill references. When the curator consolidates
-    # skill X into umbrella Y, any cron job that lists X fails to load
-    # it at run time — the scheduler skips it and the job runs without
-    # the instructions it was scheduled to follow. Rewriting the
-    # references in-place keeps scheduled jobs working across
-    # consolidation passes. Best-effort: never let a cron-module issue
-    # break the curator.
-    cron_rewrites: Dict[str, Any] = {"rewrites": [], "jobs_updated": 0, "jobs_scanned": 0}
-    try:
-        consolidated_map = {
-            e["name"]: e["into"]
-            for e in consolidated
-            if isinstance(e, dict) and e.get("name") and e.get("into")
-        }
-        pruned_names = [
-            e["name"] for e in pruned
-            if isinstance(e, dict) and e.get("name")
-        ]
-        if consolidated_map or pruned_names:
-            from cron.jobs import rewrite_skill_refs as _rewrite_cron_refs
-            cron_rewrites = _rewrite_cron_refs(
-                consolidated=consolidated_map,
-                pruned=pruned_names,
-            )
-    except Exception as e:
-        logger.debug("Curator cron skill rewrite failed: %s", e, exc_info=True)
-        cron_rewrites = {
-            "rewrites": [],
-            "jobs_updated": 0,
-            "jobs_scanned": 0,
-            "error": str(e),
-        }
-
     payload = {
         "started_at": started_at.isoformat(),
         "duration_seconds": round(elapsed_seconds, 2),
@@ -815,7 +781,6 @@ def _write_run_report(
             "consolidated_this_run": len(consolidated),
             "pruned_this_run": len(pruned),
             "state_transitions": len(transitions),
-            "cron_jobs_rewritten": int(cron_rewrites.get("jobs_updated", 0)),
             "tool_calls_total": sum(tc_counts.values()),
         },
         "tool_call_counts": tc_counts,
@@ -825,7 +790,6 @@ def _write_run_report(
         "pruned_names": [p["name"] for p in pruned],
         "added": added,
         "state_transitions": transitions,
-        "cron_rewrites": cron_rewrites,
         "llm_final": llm_meta.get("final", ""),
         "llm_summary": llm_meta.get("summary", ""),
         "llm_error": llm_meta.get("error"),
@@ -848,17 +812,6 @@ def _write_run_report(
     except Exception as e:
         logger.debug("Curator REPORT.md write failed: %s", e)
 
-    # cron_rewrites.json — only when at least one job was touched, to
-    # keep run dirs uncluttered for the common no-op case.
-    try:
-        if int(cron_rewrites.get("jobs_updated", 0)) > 0:
-            (run_dir / "cron_rewrites.json").write_text(
-                json.dumps(cron_rewrites, indent=2, ensure_ascii=False) + "\n",
-                encoding="utf-8",
-            )
-    except Exception as e:
-        logger.debug("Curator cron_rewrites.json write failed: %s", e)
-
     return run_dir
 
 
@@ -989,39 +942,6 @@ def _render_report_markdown(p: Dict[str, Any]) -> str:
             lines.append(f"- `{t.get('name')}`: {t.get('from')} → {t.get('to')}")
         lines.append("")
 
-    # Cron job rewrites — show which scheduled jobs had their skill
-    # references updated so users can audit that the auto-rewrite did
-    # the right thing. Only present when at least one job changed.
-    cron_rw = p.get("cron_rewrites") or {}
-    cron_rewrites_list = cron_rw.get("rewrites") or []
-    if cron_rewrites_list:
-        lines.append(f"### Cron job skill references rewritten ({len(cron_rewrites_list)})\n")
-        lines.append(
-            "_Cron jobs that referenced a consolidated or pruned skill were "
-            "updated in-place so they keep loading the right instructions "
-            "on their next run. See `cron_rewrites.json` for the full record._\n"
-        )
-        SHOW = 25
-        for entry in cron_rewrites_list[:SHOW]:
-            job_name = entry.get("job_name") or entry.get("job_id") or "?"
-            before = entry.get("before") or []
-            after = entry.get("after") or []
-            mapped = entry.get("mapped") or {}
-            dropped = entry.get("dropped") or []
-            lines.append(
-                f"- `{job_name}`: `{', '.join(before)}` → `{', '.join(after) or '(none)'}`"
-            )
-            for old, new in mapped.items():
-                lines.append(f"    - `{old}` → `{new}` (consolidated)")
-            for name in dropped:
-                lines.append(f"    - `{name}` dropped (pruned)")
-        if len(cron_rewrites_list) > SHOW:
-            lines.append(
-                f"- … and {len(cron_rewrites_list) - SHOW} more "
-                "(see `cron_rewrites.json`)"
-            )
-        lines.append("")
-
     # Full LLM final response
     final = (p.get("llm_final") or "").strip()
     if final:

agent/manual_compression_feedback.py

@@ -20,25 +20,25 @@ def summarize_manual_compression(
         headline = f"No changes from compression: {before_count} messages"
         if after_tokens == before_tokens:
             token_line = (
-                f"Approx request size: ~{before_tokens:,} tokens (unchanged)"
+                f"Rough transcript estimate: ~{before_tokens:,} tokens (unchanged)"
             )
         else:
             token_line = (
-                f"Approx request size: ~{before_tokens:,} → "
+                f"Rough transcript estimate: ~{before_tokens:,} → "
                 f"~{after_tokens:,} tokens"
             )
     else:
         headline = f"Compressed: {before_count} → {after_count} messages"
         token_line = (
-            f"Approx request size: ~{before_tokens:,} → "
+            f"Rough transcript estimate: ~{before_tokens:,} → "
             f"~{after_tokens:,} tokens"
         )
 
     note = None
     if not noop and after_count < before_count and after_tokens > before_tokens:
         note = (
-            "Note: fewer messages can still raise this estimate when "
-            "compression rewrites the transcript into denser summaries."
+            "Note: fewer messages can still raise this rough transcript estimate "
+            "when compression rewrites the transcript into denser summaries."
         )
 
     return {

agent/moonshot_schema.py

@@ -15,6 +15,16 @@ and MoonshotAI/kimi-cli#1595:
 2. When ``anyOf`` is used, ``type`` must be on the ``anyOf`` children, not
    the parent.  Presence of both causes "type should be defined in anyOf
    items instead of the parent schema".
+3. ``$ref`` nodes may not carry sibling keywords.  Moonshot expands the
+   reference before validation and then rejects the node if sibling keys
+   like ``description`` remain on the same node as ``$ref``.  Strip every
+   sibling from ``$ref`` nodes so only ``{"$ref": "..."}`` survives.
+   (Ported from anomalyco/opencode#24730.)
+4. ``items`` may not be a tuple-style array (``items: [schemaA, schemaB]``
+   for positional element schemas).  Moonshot's schema engine requires a
+   single object schema applied to every array element.  Collapse tuple
+   ``items`` to the first element schema (or ``{}`` if the tuple is empty).
+   (Ported from anomalyco/opencode#24730.)
 
 The ``#/definitions/...`` → ``#/$defs/...`` rewrite for draft-07 refs is
 handled separately in ``tools/mcp_tool._normalize_mcp_input_schema`` so it
@@ -66,6 +76,16 @@ def _repair_schema(node: Any, is_schema: bool = True) -> Any:
             }
         elif key in _SCHEMA_LIST_KEYS and isinstance(value, list):
             repaired[key] = [_repair_schema(v, is_schema=True) for v in value]
+        elif key == "items" and isinstance(value, list):
+            # Rule 4: tuple-style ``items`` arrays (positional element
+            # schemas) are not accepted by Moonshot.  Collapse to the
+            # first element schema if present, else to ``{}``.  This
+            # matches opencode's behaviour for moonshotai / kimi models.
+            first = value[0] if value else {}
+            if isinstance(first, dict):
+                repaired[key] = _repair_schema(first, is_schema=True)
+            else:
+                repaired[key] = first
         elif key in _SCHEMA_NODE_KEYS:
             # items / not / additionalProperties: single nested schema.
             # additionalProperties can also be a bool — leave those alone.
@@ -81,56 +101,17 @@ def _repair_schema(node: Any, is_schema: bool = True) -> Any:
         return repaired
 
     # Rule 2: when anyOf is present, type belongs only on the children.
-    # Additionally, Moonshot rejects null-type branches inside anyOf
-    # (enum value (<nil>) does not match any type in [string]).
-    # Collapse the anyOf to the first non-null branch and infer its type.
     if "anyOf" in repaired and isinstance(repaired["anyOf"], list):
         repaired.pop("type", None)
-        non_null = [b for b in repaired["anyOf"]
-                    if isinstance(b, dict) and b.get("type") != "null"]
-        if non_null and len(non_null) < len(repaired["anyOf"]):
-            # Drop the anyOf wrapper — keep only the non-null branch.
-            # If there's a single non-null branch, promote it and fall
-            # through to Rules 1/3 so nullable/enum cleanup still applies
-            # to the merged node.
-            if len(non_null) == 1:
-                merge = {k: v for k, v in repaired.items() if k != "anyOf"}
-                merge.update(non_null[0])
-                repaired = merge
-            else:
-                repaired["anyOf"] = non_null
-                return repaired
-        else:
-            # Nothing to collapse — parent type stripped, children already
-            # repaired by the recursive walk above.
-            return repaired
+        return repaired
 
-    # Moonshot also rejects non-standard keywords like ``nullable`` on
-    # parameter schemas — strip it.
-    repaired.pop("nullable", None)
-
-    # Rule 1: property schemas without type need one.  $ref nodes are exempt
-    # — their type comes from the referenced definition.
-    # Fill missing type BEFORE Rule 3 so enum cleanup can check the type.
-    if "$ref" not in repaired:
-        repaired = _fill_missing_type(repaired)
-
-    # Rule 3: Moonshot rejects null/empty-string values inside enum arrays
-    # when the parent type is a scalar (string, integer, etc.).  The error:
-    #   "enum value (<nil>) does not match any type in [string]"
-    # Strip null and empty-string from enum values, and if the enum becomes
-    # empty, drop it entirely.
-    if "enum" in repaired and isinstance(repaired["enum"], list):
-        node_type = repaired.get("type")
-        if node_type in ("string", "integer", "number", "boolean"):
-            cleaned = [v for v in repaired["enum"]
-                       if v is not None and v != ""]
-            if cleaned:
-                repaired["enum"] = cleaned
-            else:
-                repaired.pop("enum")
-
-    return repaired
+    # Rule 3: $ref nodes must not have sibling keywords.  Strip everything
+    # except $ref itself so Moonshot's validator (which expands the ref
+    # before checking) doesn't reject the node for redundant keys like
+    # ``description`` / ``type`` / ``default`` appearing alongside $ref.
+    if "$ref" in repaired:
+        return {"$ref": repaired["$ref"]}
+    return _fill_missing_type(repaired)
 
 
 def _fill_missing_type(node: Dict[str, Any]) -> Dict[str, Any]:

agent/tool_guardrails.py

@@ -1,455 +0,0 @@
-"""Pure tool-call loop guardrail primitives.
-
-The controller in this module is intentionally side-effect free: it tracks
-per-turn tool-call observations and returns decisions. Runtime code owns whether
-those decisions become warning guidance, synthetic tool results, or controlled
-turn halts.
-"""
-
-from __future__ import annotations
-
-import hashlib
-import json
-from dataclasses import dataclass, field
-from typing import Any, Mapping
-
-from utils import safe_json_loads
-
-
-IDEMPOTENT_TOOL_NAMES = frozenset(
-    {
-        "read_file",
-        "search_files",
-        "web_search",
-        "web_extract",
-        "session_search",
-        "browser_snapshot",
-        "browser_console",
-        "browser_get_images",
-        "mcp_filesystem_read_file",
-        "mcp_filesystem_read_text_file",
-        "mcp_filesystem_read_multiple_files",
-        "mcp_filesystem_list_directory",
-        "mcp_filesystem_list_directory_with_sizes",
-        "mcp_filesystem_directory_tree",
-        "mcp_filesystem_get_file_info",
-        "mcp_filesystem_search_files",
-    }
-)
-
-MUTATING_TOOL_NAMES = frozenset(
-    {
-        "terminal",
-        "execute_code",
-        "write_file",
-        "patch",
-        "todo",
-        "memory",
-        "skill_manage",
-        "browser_click",
-        "browser_type",
-        "browser_press",
-        "browser_scroll",
-        "browser_navigate",
-        "send_message",
-        "cronjob",
-        "delegate_task",
-        "process",
-    }
-)
-
-
-@dataclass(frozen=True)
-class ToolCallGuardrailConfig:
-    """Thresholds for per-turn tool-call loop detection.
-
-    Warnings are enabled by default and never prevent tool execution. Hard stops
-    are explicit opt-in so interactive CLI/TUI sessions get a gentle nudge unless
-    the user enables circuit-breaker behavior in config.yaml.
-    """
-
-    warnings_enabled: bool = True
-    hard_stop_enabled: bool = False
-    exact_failure_warn_after: int = 2
-    exact_failure_block_after: int = 5
-    same_tool_failure_warn_after: int = 3
-    same_tool_failure_halt_after: int = 8
-    no_progress_warn_after: int = 2
-    no_progress_block_after: int = 5
-    idempotent_tools: frozenset[str] = field(default_factory=lambda: IDEMPOTENT_TOOL_NAMES)
-    mutating_tools: frozenset[str] = field(default_factory=lambda: MUTATING_TOOL_NAMES)
-
-    @classmethod
-    def from_mapping(cls, data: Mapping[str, Any] | None) -> "ToolCallGuardrailConfig":
-        """Build config from the `tool_loop_guardrails` config.yaml section."""
-        if not isinstance(data, Mapping):
-            return cls()
-
-        warn_after = data.get("warn_after")
-        if not isinstance(warn_after, Mapping):
-            warn_after = {}
-        hard_stop_after = data.get("hard_stop_after")
-        if not isinstance(hard_stop_after, Mapping):
-            hard_stop_after = {}
-
-        defaults = cls()
-        return cls(
-            warnings_enabled=_as_bool(data.get("warnings_enabled"), defaults.warnings_enabled),
-            hard_stop_enabled=_as_bool(data.get("hard_stop_enabled"), defaults.hard_stop_enabled),
-            exact_failure_warn_after=_positive_int(
-                warn_after.get("exact_failure", data.get("exact_failure_warn_after")),
-                defaults.exact_failure_warn_after,
-            ),
-            same_tool_failure_warn_after=_positive_int(
-                warn_after.get("same_tool_failure", data.get("same_tool_failure_warn_after")),
-                defaults.same_tool_failure_warn_after,
-            ),
-            no_progress_warn_after=_positive_int(
-                warn_after.get("idempotent_no_progress", data.get("no_progress_warn_after")),
-                defaults.no_progress_warn_after,
-            ),
-            exact_failure_block_after=_positive_int(
-                hard_stop_after.get("exact_failure", data.get("exact_failure_block_after")),
-                defaults.exact_failure_block_after,
-            ),
-            same_tool_failure_halt_after=_positive_int(
-                hard_stop_after.get("same_tool_failure", data.get("same_tool_failure_halt_after")),
-                defaults.same_tool_failure_halt_after,
-            ),
-            no_progress_block_after=_positive_int(
-                hard_stop_after.get("idempotent_no_progress", data.get("no_progress_block_after")),
-                defaults.no_progress_block_after,
-            ),
-        )
-
-
-@dataclass(frozen=True)
-class ToolCallSignature:
-    """Stable, non-reversible identity for a tool name plus canonical args."""
-
-    tool_name: str
-    args_hash: str
-
-    @classmethod
-    def from_call(cls, tool_name: str, args: Mapping[str, Any] | None) -> "ToolCallSignature":
-        canonical = canonical_tool_args(args or {})
-        return cls(tool_name=tool_name, args_hash=_sha256(canonical))
-
-    def to_metadata(self) -> dict[str, str]:
-        """Return public metadata without raw argument values."""
-        return {"tool_name": self.tool_name, "args_hash": self.args_hash}
-
-
-@dataclass(frozen=True)
-class ToolGuardrailDecision:
-    """Decision returned by the tool-call guardrail controller."""
-
-    action: str = "allow"  # allow | warn | block | halt
-    code: str = "allow"
-    message: str = ""
-    tool_name: str = ""
-    count: int = 0
-    signature: ToolCallSignature | None = None
-
-    @property
-    def allows_execution(self) -> bool:
-        return self.action in {"allow", "warn"}
-
-    @property
-    def should_halt(self) -> bool:
-        return self.action in {"block", "halt"}
-
-    def to_metadata(self) -> dict[str, Any]:
-        data: dict[str, Any] = {
-            "action": self.action,
-            "code": self.code,
-            "message": self.message,
-            "tool_name": self.tool_name,
-            "count": self.count,
-        }
-        if self.signature is not None:
-            data["signature"] = self.signature.to_metadata()
-        return data
-
-
-def canonical_tool_args(args: Mapping[str, Any]) -> str:
-    """Return sorted compact JSON for parsed tool arguments."""
-    if not isinstance(args, Mapping):
-        raise TypeError(f"tool args must be a mapping, got {type(args).__name__}")
-    return json.dumps(
-        args,
-        ensure_ascii=False,
-        sort_keys=True,
-        separators=(",", ":"),
-        default=str,
-    )
-
-
-def classify_tool_failure(tool_name: str, result: str | None) -> tuple[bool, str]:
-    """Safety-fallback classifier used only when callers don't pass ``failed``.
-
-    Mirrors ``agent.display._detect_tool_failure`` exactly so the guardrail
-    never disagrees with the CLI's user-visible ``[error]`` tag. Production
-    callers in ``run_agent.py`` always pass an explicit ``failed=`` derived
-    from ``_detect_tool_failure``; this function exists so standalone callers
-    (tests, tooling) still get consistent behavior.
-    """
-    if result is None:
-        return False, ""
-
-    if tool_name == "terminal":
-        data = safe_json_loads(result)
-        if isinstance(data, dict):
-            exit_code = data.get("exit_code")
-            if exit_code is not None and exit_code != 0:
-                return True, f" [exit {exit_code}]"
-        return False, ""
-
-    if tool_name == "memory":
-        data = safe_json_loads(result)
-        if isinstance(data, dict):
-            if data.get("success") is False and "exceed the limit" in data.get("error", ""):
-                return True, " [full]"
-
-    lower = result[:500].lower()
-    if '"error"' in lower or '"failed"' in lower or result.startswith("Error"):
-        return True, " [error]"
-
-    return False, ""
-
-
-class ToolCallGuardrailController:
-    """Per-turn controller for repeated failed/non-progressing tool calls."""
-
-    def __init__(self, config: ToolCallGuardrailConfig | None = None):
-        self.config = config or ToolCallGuardrailConfig()
-        self.reset_for_turn()
-
-    def reset_for_turn(self) -> None:
-        self._exact_failure_counts: dict[ToolCallSignature, int] = {}
-        self._same_tool_failure_counts: dict[str, int] = {}
-        self._no_progress: dict[ToolCallSignature, tuple[str, int]] = {}
-        self._halt_decision: ToolGuardrailDecision | None = None
-
-    @property
-    def halt_decision(self) -> ToolGuardrailDecision | None:
-        return self._halt_decision
-
-    def before_call(self, tool_name: str, args: Mapping[str, Any] | None) -> ToolGuardrailDecision:
-        signature = ToolCallSignature.from_call(tool_name, _coerce_args(args))
-        if not self.config.hard_stop_enabled:
-            return ToolGuardrailDecision(tool_name=tool_name, signature=signature)
-
-        exact_count = self._exact_failure_counts.get(signature, 0)
-        if exact_count >= self.config.exact_failure_block_after:
-            decision = ToolGuardrailDecision(
-                action="block",
-                code="repeated_exact_failure_block",
-                message=(
-                    f"Blocked {tool_name}: the same tool call failed {exact_count} "
-                    "times with identical arguments. Stop retrying it unchanged; "
-                    "change strategy or explain the blocker."
-                ),
-                tool_name=tool_name,
-                count=exact_count,
-                signature=signature,
-            )
-            self._halt_decision = decision
-            return decision
-
-        if self._is_idempotent(tool_name):
-            record = self._no_progress.get(signature)
-            if record is not None:
-                _result_hash, repeat_count = record
-                if repeat_count >= self.config.no_progress_block_after:
-                    decision = ToolGuardrailDecision(
-                        action="block",
-                        code="idempotent_no_progress_block",
-                        message=(
-                            f"Blocked {tool_name}: this read-only call returned the same "
-                            f"result {repeat_count} times. Stop repeating it unchanged; "
-                            "use the result already provided or try a different query."
-                        ),
-                        tool_name=tool_name,
-                        count=repeat_count,
-                        signature=signature,
-                    )
-                    self._halt_decision = decision
-                    return decision
-
-        return ToolGuardrailDecision(tool_name=tool_name, signature=signature)
-
-    def after_call(
-        self,
-        tool_name: str,
-        args: Mapping[str, Any] | None,
-        result: str | None,
-        *,
-        failed: bool | None = None,
-    ) -> ToolGuardrailDecision:
-        args = _coerce_args(args)
-        signature = ToolCallSignature.from_call(tool_name, args)
-        if failed is None:
-            failed, _ = classify_tool_failure(tool_name, result)
-
-        if failed:
-            exact_count = self._exact_failure_counts.get(signature, 0) + 1
-            self._exact_failure_counts[signature] = exact_count
-            self._no_progress.pop(signature, None)
-
-            same_count = self._same_tool_failure_counts.get(tool_name, 0) + 1
-            self._same_tool_failure_counts[tool_name] = same_count
-
-            if self.config.hard_stop_enabled and same_count >= self.config.same_tool_failure_halt_after:
-                decision = ToolGuardrailDecision(
-                    action="halt",
-                    code="same_tool_failure_halt",
-                    message=(
-                        f"Stopped {tool_name}: it failed {same_count} times this turn. "
-                        "Stop retrying the same failing tool path and choose a different approach."
-                    ),
-                    tool_name=tool_name,
-                    count=same_count,
-                    signature=signature,
-                )
-                self._halt_decision = decision
-                return decision
-
-            if self.config.warnings_enabled and exact_count >= self.config.exact_failure_warn_after:
-                return ToolGuardrailDecision(
-                    action="warn",
-                    code="repeated_exact_failure_warning",
-                    message=(
-                        f"{tool_name} has failed {exact_count} times with identical arguments. "
-                        "This looks like a loop; inspect the error and change strategy "
-                        "instead of retrying it unchanged."
-                    ),
-                    tool_name=tool_name,
-                    count=exact_count,
-                    signature=signature,
-                )
-
-            if self.config.warnings_enabled and same_count >= self.config.same_tool_failure_warn_after:
-                return ToolGuardrailDecision(
-                    action="warn",
-                    code="same_tool_failure_warning",
-                    message=(
-                        f"{tool_name} has failed {same_count} times this turn. "
-                        "This looks like a loop; change approach before retrying."
-                    ),
-                    tool_name=tool_name,
-                    count=same_count,
-                    signature=signature,
-                )
-
-            return ToolGuardrailDecision(tool_name=tool_name, count=exact_count, signature=signature)
-
-        self._exact_failure_counts.pop(signature, None)
-        self._same_tool_failure_counts.pop(tool_name, None)
-
-        if not self._is_idempotent(tool_name):
-            self._no_progress.pop(signature, None)
-            return ToolGuardrailDecision(tool_name=tool_name, signature=signature)
-
-        result_hash = _result_hash(result)
-        previous = self._no_progress.get(signature)
-        repeat_count = 1
-        if previous is not None and previous[0] == result_hash:
-            repeat_count = previous[1] + 1
-        self._no_progress[signature] = (result_hash, repeat_count)
-
-        if self.config.warnings_enabled and repeat_count >= self.config.no_progress_warn_after:
-            return ToolGuardrailDecision(
-                action="warn",
-                code="idempotent_no_progress_warning",
-                message=(
-                    f"{tool_name} returned the same result {repeat_count} times. "
-                    "Use the result already provided or change the query instead of "
-                    "repeating it unchanged."
-                ),
-                tool_name=tool_name,
-                count=repeat_count,
-                signature=signature,
-            )
-
-        return ToolGuardrailDecision(tool_name=tool_name, count=repeat_count, signature=signature)
-
-    def _is_idempotent(self, tool_name: str) -> bool:
-        if tool_name in self.config.mutating_tools:
-            return False
-        return tool_name in self.config.idempotent_tools
-
-
-def toolguard_synthetic_result(decision: ToolGuardrailDecision) -> str:
-    """Build a synthetic role=tool content string for a blocked tool call."""
-    return json.dumps(
-        {
-            "error": decision.message,
-            "guardrail": decision.to_metadata(),
-        },
-        ensure_ascii=False,
-    )
-
-
-def append_toolguard_guidance(result: str, decision: ToolGuardrailDecision) -> str:
-    """Append runtime guidance to the current tool result content."""
-    if decision.action not in {"warn", "halt"} or not decision.message:
-        return result
-    label = "Tool loop hard stop" if decision.action == "halt" else "Tool loop warning"
-    suffix = (
-        f"\n\n[{label}: "
-        f"{decision.code}; count={decision.count}; {decision.message}]"
-    )
-    return (result or "") + suffix
-
-
-def _coerce_args(args: Mapping[str, Any] | None) -> Mapping[str, Any]:
-    return args if isinstance(args, Mapping) else {}
-
-
-def _result_hash(result: str | None) -> str:
-    parsed = safe_json_loads(result or "")
-    if parsed is not None:
-        try:
-            canonical = json.dumps(
-                parsed,
-                ensure_ascii=False,
-                sort_keys=True,
-                separators=(",", ":"),
-                default=str,
-            )
-        except TypeError:
-            canonical = str(parsed)
-    else:
-        canonical = result or ""
-    return _sha256(canonical)
-
-
-def _as_bool(value: Any, default: bool) -> bool:
-    if value is None:
-        return default
-    if isinstance(value, bool):
-        return value
-    if isinstance(value, (int, float)):
-        return bool(value)
-    if isinstance(value, str):
-        lowered = value.strip().lower()
-        if lowered in {"1", "true", "yes", "on", "enabled"}:
-            return True
-        if lowered in {"0", "false", "no", "off", "disabled"}:
-            return False
-    return default
-
-
-def _positive_int(value: Any, default: int) -> int:
-    if value is None:
-        return default
-    try:
-        parsed = int(value)
-    except (TypeError, ValueError):
-        return default
-    return parsed if parsed >= 1 else default
-
-
-def _sha256(value: str) -> str:
-    return hashlib.sha256(value.encode("utf-8")).hexdigest()

cli-config.yaml.example

@@ -289,25 +289,6 @@ browser:
   # after this period of no activity between agent loops (default: 120 = 2 minutes)
   inactivity_timeout: 120
 
-# =============================================================================
-# Tool Loop Guardrails
-# =============================================================================
-# Soft warnings are enabled by default. They append guidance to repeated failed
-# or non-progressing tool results but still let the tool execute. Hard stops are
-# opt-in circuit breakers for autonomous/cron sessions where stopping a loop is
-# preferable to spending the full iteration budget.
-tool_loop_guardrails:
-  warnings_enabled: true
-  hard_stop_enabled: false
-  warn_after:
-    exact_failure: 2
-    same_tool_failure: 3
-    idempotent_no_progress: 2
-  hard_stop_after:
-    exact_failure: 5
-    same_tool_failure: 8
-    idempotent_no_progress: 5
-
 # =============================================================================
 # Context Compression (Auto-shrinks long conversations)
 # =============================================================================

cli.py

@@ -15,6 +15,7 @@ Usage:
 
 import logging
 import os
+import re
 import shutil
 import sys
 import json
@@ -85,7 +86,7 @@ from hermes_cli.browser_connect import (
     try_launch_chrome_debug,
 )
 from hermes_cli.env_loader import load_hermes_dotenv
-from utils import base_url_host_matches, is_truthy_value
+from utils import base_url_host_matches
 
 _hermes_home = get_hermes_home()
 _project_env = Path(__file__).parent / '.env'
@@ -599,7 +600,6 @@ def load_cli_config() -> Dict[str, Any]:
 # Load configuration at module startup
 CLI_CONFIG = load_cli_config()
 
-
 # Initialize centralized logging early — agent.log + errors.log in ~/.hermes/logs/.
 # This ensures CLI sessions produce a log trail even before AIAgent is instantiated.
 try:
@@ -934,20 +934,6 @@ def _run_state_db_auto_maintenance(session_db) -> None:
     try:
         from hermes_cli.config import load_config as _load_full_config
         from hermes_constants import get_hermes_home as _get_hermes_home
-        _hermes_home_maint = _get_hermes_home()
-
-        # One-time prune of empty TUI ghost sessions.
-        try:
-            if not session_db.get_meta("ghost_session_prune_v1"):
-                pruned = session_db.prune_empty_ghost_sessions(
-                    sessions_dir=_hermes_home_maint / "sessions"
-                )
-                session_db.set_meta("ghost_session_prune_v1", "1")
-                if pruned:
-                    logger.info("Pruned %d empty TUI ghost sessions", pruned)
-        except Exception as _prune_exc:
-            logger.debug("Ghost session prune skipped: %s", _prune_exc)
-
         cfg = (_load_full_config().get("sessions") or {})
         if not cfg.get("auto_prune", False):
             return
@@ -955,7 +941,7 @@ def _run_state_db_auto_maintenance(session_db) -> None:
             retention_days=int(cfg.get("retention_days", 90)),
             min_interval_hours=int(cfg.get("min_interval_hours", 24)),
             vacuum=bool(cfg.get("vacuum_after_prune", True)),
-            sessions_dir=_hermes_home_maint / "sessions",
+            sessions_dir=_get_hermes_home() / "sessions",
         )
     except Exception as exc:
         logger.debug("state.db auto-maintenance skipped: %s", exc)
@@ -2132,8 +2118,6 @@ class HermesCLI:
         
         # Parse and validate toolsets
         self.enabled_toolsets = toolsets
-        self.disabled_toolsets = CLI_CONFIG["agent"].get("disabled_toolsets") or []
-
         if toolsets and "all" not in toolsets and "*" not in toolsets:
             # Validate each toolset — MCP server names are resolved via
             # live registry aliases (registered during discover_mcp_tools),
@@ -3584,7 +3568,6 @@ class HermesCLI:
                 credential_pool=runtime.get("credential_pool"),
                 max_iterations=self.max_turns,
                 enabled_toolsets=self.enabled_toolsets,
-                disabled_toolsets=self.disabled_toolsets,
                 verbose_logging=self.verbose,
                 quiet_mode=not self.verbose,
                 ephemeral_system_prompt=self.system_prompt if self.system_prompt else None,
@@ -3632,18 +3615,14 @@ class HermesCLI:
                 tuple(runtime.get("args") or ()),
             )
 
-            # Force-create DB row on /title intent, then apply title.
-            if self._pending_title and self._session_db and self.agent:
+            if self._pending_title and self._session_db:
                 try:
-                    self.agent._ensure_db_session()
-                    if self.agent._session_db_created:
-                        self._session_db.set_session_title(self.session_id, self._pending_title)
-                        _cprint(f"  Session title applied: {self._pending_title}")
-                        self._pending_title = None
-                    # else: row creation failed transiently — keep _pending_title for retry
+                    self._session_db.set_session_title(self.session_id, self._pending_title)
+                    _cprint(f"  Session title applied: {self._pending_title}")
+                    self._pending_title = None
                 except (ValueError, Exception) as e:
                     _cprint(f"  Could not apply pending title: {e}")
-                    # Keep _pending_title so it can be retried after row creation succeeds
+                    self._pending_title = None
             return True
         except Exception as e:
             ChatConsole().print(f"[bold red]Failed to initialize agent: {e}[/]")
@@ -4971,7 +4950,6 @@ class HermesCLI:
 
             if self._session_db:
                 try:
-                    self.agent._session_db_created = False
                     self._session_db.create_session(
                         session_id=self.session_id,
                         source=os.environ.get("HERMES_SESSION_SOURCE", "cli"),
@@ -4981,7 +4959,6 @@ class HermesCLI:
                             "reasoning_config": self.reasoning_config,
                         },
                     )
-                    self.agent._session_db_created = True
                 except Exception:
                     pass
             # Notify memory providers that session_id rotated to a fresh
@@ -6560,8 +6537,6 @@ class HermesCLI:
                 # No active run — treat as a normal next-turn message.
                 self._pending_input.put(payload)
                 _cprint(f"  No agent running; queued as next turn: {payload[:80]}{'...' if len(payload) > 80 else ''}")
-        elif canonical == "goal":
-            self._handle_goal_command(cmd_original)
         elif canonical == "skin":
             self._handle_skin_command(cmd_original)
         elif canonical == "voice":
@@ -6607,17 +6582,12 @@ class HermesCLI:
                     self._console_print(f"[bold red]Quick command '{base_cmd}' has unsupported type (supported: 'exec', 'alias')[/]")
             # Check for plugin-registered slash commands
             elif base_cmd.lstrip("/") in _get_plugin_cmd_handler_names():
-                from hermes_cli.plugins import (
-                    get_plugin_command_handler,
-                    resolve_plugin_command_result,
-                )
+                from hermes_cli.plugins import get_plugin_command_handler
                 plugin_handler = get_plugin_command_handler(base_cmd.lstrip("/"))
                 if plugin_handler:
                     user_args = cmd_original[len(base_cmd):].strip()
                     try:
-                        result = resolve_plugin_command_result(
-                            plugin_handler(user_args)
-                        )
+                        result = plugin_handler(user_args)
                         if result:
                             _cprint(str(result))
                     except Exception as e:
@@ -7042,166 +7012,6 @@ class HermesCLI:
             print("   status       Show current browser mode")
             print()
 
-    # ────────────────────────────────────────────────────────────────
-    # /goal — persistent cross-turn goals (Ralph-style loop)
-    # ────────────────────────────────────────────────────────────────
-    def _get_goal_manager(self):
-        """Return the GoalManager bound to the current session_id.
-
-        Cached on ``self._goal_manager`` and rebound lazily when
-        ``session_id`` changes (e.g. after /new or a compression-driven
-        session split).
-        """
-        try:
-            from hermes_cli.goals import GoalManager
-            from hermes_cli.config import load_config
-        except Exception as exc:
-            logging.debug("goal manager unavailable: %s", exc)
-            return None
-
-        sid = getattr(self, "session_id", None) or ""
-        if not sid:
-            return None
-
-        existing = getattr(self, "_goal_manager", None)
-        if existing is not None and getattr(existing, "session_id", None) == sid:
-            return existing
-
-        try:
-            cfg = load_config() or {}
-            goals_cfg = cfg.get("goals") or {}
-            max_turns = int(goals_cfg.get("max_turns", 20) or 20)
-        except Exception:
-            max_turns = 20
-
-        mgr = GoalManager(session_id=sid, default_max_turns=max_turns)
-        self._goal_manager = mgr
-        return mgr
-
-    def _handle_goal_command(self, cmd: str) -> None:
-        """Dispatch /goal subcommands: set / status / pause / resume / clear."""
-        parts = (cmd or "").strip().split(None, 1)
-        arg = parts[1].strip() if len(parts) > 1 else ""
-
-        mgr = self._get_goal_manager()
-        if mgr is None:
-            _cprint(f"  {_DIM}Goals unavailable (no active session).{_RST}")
-            return
-
-        lower = arg.lower()
-
-        # Bare /goal or /goal status → show current state
-        if not arg or lower == "status":
-            _cprint(f"  {mgr.status_line()}")
-            return
-
-        if lower == "pause":
-            state = mgr.pause(reason="user-paused")
-            if state is None:
-                _cprint(f"  {_DIM}No goal set.{_RST}")
-            else:
-                _cprint(f"  ⏸ Goal paused: {state.goal}")
-            return
-
-        if lower == "resume":
-            state = mgr.resume()
-            if state is None:
-                _cprint(f"  {_DIM}No goal to resume.{_RST}")
-            else:
-                _cprint(f"  ▶ Goal resumed: {state.goal}")
-                _cprint(
-                    f"  {_DIM}Send any message (or press Enter on an empty prompt "
-                    f"is a no-op; type 'continue' to kick it off).{_RST}"
-                )
-            return
-
-        if lower in ("clear", "stop", "done"):
-            had = mgr.has_goal()
-            mgr.clear()
-            if had:
-                _cprint("  ✓ Goal cleared.")
-            else:
-                _cprint(f"  {_DIM}No active goal.{_RST}")
-            return
-
-        # Otherwise treat the arg as the goal text.
-        try:
-            state = mgr.set(arg)
-        except ValueError as exc:
-            _cprint(f"  Invalid goal: {exc}")
-            return
-
-        _cprint(f"  ⊙ Goal set ({state.max_turns}-turn budget): {state.goal}")
-        _cprint(
-            f"  {_DIM}After each turn, a judge model will check if the goal is done. "
-            f"Hermes keeps working until it is, you pause/clear it, or the budget is "
-            f"exhausted. Use /goal status, /goal pause, /goal resume, /goal clear.{_RST}"
-        )
-        # Kick the loop off immediately so the user doesn't have to send a
-        # separate message after setting the goal.
-        try:
-            self._pending_input.put(state.goal)
-        except Exception:
-            pass
-
-    def _maybe_continue_goal_after_turn(self) -> None:
-        """Hook run after every CLI turn. Judges + maybe re-queues.
-
-        Safe to call when no goal is set — returns quickly.
-
-        Preemption is automatic: if a real user message is already in
-        ``_pending_input`` we skip judging (the user's new input takes
-        priority and we'll re-judge after that turn). If judge says done,
-        mark it done and tell the user. If judge says continue and we're
-        under budget, push the continuation prompt onto the queue.
-        """
-        mgr = self._get_goal_manager()
-        if mgr is None or not mgr.is_active():
-            return
-
-        # If a real user message is already queued, don't inject a
-        # continuation prompt on top — let the user's turn go first.
-        try:
-            if getattr(self, "_pending_input", None) is not None \
-                    and not self._pending_input.empty():
-                return
-        except Exception:
-            pass
-
-        # Extract the agent's final response for this turn.
-        last_response = ""
-        try:
-            hist = self.conversation_history or []
-            for msg in reversed(hist):
-                if msg.get("role") == "assistant":
-                    content = msg.get("content", "")
-                    if isinstance(content, list):
-                        # Multimodal content — flatten text parts.
-                        parts = [
-                            p.get("text", "")
-                            for p in content
-                            if isinstance(p, dict) and p.get("type") in ("text", "output_text")
-                        ]
-                        last_response = "\n".join(t for t in parts if t)
-                    else:
-                        last_response = str(content or "")
-                    break
-        except Exception:
-            last_response = ""
-
-        decision = mgr.evaluate_after_turn(last_response, user_initiated=True)
-        msg = decision.get("message") or ""
-        if msg:
-            _cprint(f"  {msg}")
-
-        if decision.get("should_continue"):
-            prompt = decision.get("continuation_prompt")
-            if prompt:
-                try:
-                    self._pending_input.put(prompt)
-                except Exception as exc:
-                    logging.debug("goal continuation enqueue failed: %s", exc)
-
     def _handle_skin_command(self, cmd: str):
         """Handle /skin [name] — show or change the display skin."""
         try:
@@ -7328,7 +7138,7 @@ class HermesCLI:
         import os
         from hermes_cli.colors import Colors as _Colors
 
-        current = is_truthy_value(os.environ.get("HERMES_YOLO_MODE"))
+        current = bool(os.environ.get("HERMES_YOLO_MODE"))
         if current:
             os.environ.pop("HERMES_YOLO_MODE", None)
             _cprint(
@@ -7525,20 +7335,10 @@ class HermesCLI:
         original_count = len(self.conversation_history)
         with self._busy_command("Compressing context..."):
             try:
-                from agent.model_metadata import estimate_request_tokens_rough
+                from agent.model_metadata import estimate_messages_tokens_rough
                 from agent.manual_compression_feedback import summarize_manual_compression
                 original_history = list(self.conversation_history)
-                # Include system prompt + tool schemas in the estimate —
-                # a transcript-only number understates real request pressure
-                # and can even appear to grow after compression because a
-                # dense handoff summary replaces many short turns (#6217).
-                _sys_prompt = getattr(self.agent, "_cached_system_prompt", "") or ""
-                _tools = getattr(self.agent, "tools", None) or None
-                approx_tokens = estimate_request_tokens_rough(
-                    original_history,
-                    system_prompt=_sys_prompt,
-                    tools=_tools,
-                )
+                approx_tokens = estimate_messages_tokens_rough(original_history)
                 if focus_topic:
                     print(f"🗜️  Compressing {original_count} messages (~{approx_tokens:,} tokens), "
                           f"focus: \"{focus_topic}\"...")
@@ -7570,11 +7370,7 @@ class HermesCLI:
                 ):
                     self.session_id = self.agent.session_id
                     self._pending_title = None
-                new_tokens = estimate_request_tokens_rough(
-                    self.conversation_history,
-                    system_prompt=_sys_prompt,
-                    tools=_tools,
-                )
+                new_tokens = estimate_messages_tokens_rough(self.conversation_history)
                 summary = summarize_manual_compression(
                     original_history,
                     self.conversation_history,
@@ -11540,17 +11336,6 @@ class HermesCLI:
 
                         app.invalidate()  # Refresh status line
 
-                        # Goal continuation: if a standing goal is active, ask
-                        # the judge whether the turn satisfied it. If not, and
-                        # there's no real user message already queued, push the
-                        # continuation prompt back into _pending_input so the
-                        # next loop iteration picks it up naturally (and any
-                        # user input that arrives in between still preempts).
-                        try:
-                            self._maybe_continue_goal_after_turn()
-                        except Exception as _goal_exc:
-                            logging.debug("goal continuation hook failed: %s", _goal_exc)
-
                         # Continuous voice: auto-restart recording after agent responds.
                         # Dispatch to a daemon thread so play_beep (sd.wait) and
                         # AudioRecorder.start (lock acquire) never block process_loop —

cron/jobs.py

@@ -882,121 +882,3 @@ def save_job_output(job_id: str, output: str):
         raise
     
     return output_file
-
-
-# =============================================================================
-# Skill reference rewriting (curator integration)
-# =============================================================================
-
-def rewrite_skill_refs(
-    consolidated: Optional[Dict[str, str]] = None,
-    pruned: Optional[List[str]] = None,
-) -> Dict[str, Any]:
-    """Rewrite cron job skill references after a curator consolidation pass.
-
-    When the curator consolidates a skill X into umbrella Y (or archives X
-    as pruned), any cron job that lists ``X`` in its ``skills`` field will
-    fail to load ``X`` at run time — the scheduler logs a warning and
-    skips the skill, so the job runs without the instructions it was
-    scheduled to follow. See cron/scheduler.py where ``skill_view`` is
-    called per skill name.
-
-    This function repairs cron jobs in-place:
-
-    - A skill listed in ``consolidated`` is replaced with its umbrella
-      target (the ``into`` value). If the umbrella is already in the
-      job's skill list, the stale name is dropped without duplication.
-    - A skill listed in ``pruned`` is dropped outright — there is no
-      forwarding target.
-    - Ordering and other skills in the list are preserved.
-    - The legacy ``skill`` field is realigned via ``_apply_skill_fields``.
-
-    Args:
-        consolidated: mapping of ``old_skill_name -> umbrella_skill_name``.
-        pruned: list of skill names that were archived with no forwarding
-            target.
-
-    Returns a report dict::
-
-        {
-            "rewrites": [
-                {
-                    "job_id": ...,
-                    "job_name": ...,
-                    "before": [...],
-                    "after": [...],
-                    "mapped": {"old": "new", ...},
-                    "dropped": ["old", ...],
-                },
-                ...
-            ],
-            "jobs_updated": N,
-            "jobs_scanned": M,
-        }
-
-    Best-effort: exceptions from loading/saving propagate to the caller so
-    tests can assert behaviour; the curator invocation site wraps this
-    call in a try/except so a failure here never breaks the curator.
-    """
-    consolidated = dict(consolidated or {})
-    pruned_set = set(pruned or [])
-    # A skill listed in both wins as "consolidated" — it has a target,
-    # which is the more useful of the two outcomes.
-    pruned_set -= set(consolidated.keys())
-
-    if not consolidated and not pruned_set:
-        return {"rewrites": [], "jobs_updated": 0, "jobs_scanned": 0}
-
-    with _jobs_file_lock:
-        jobs = load_jobs()
-        rewrites: List[Dict[str, Any]] = []
-        changed = False
-
-        for job in jobs:
-            skills_before = _normalize_skill_list(job.get("skill"), job.get("skills"))
-            if not skills_before:
-                continue
-
-            mapped: Dict[str, str] = {}
-            dropped: List[str] = []
-            new_skills: List[str] = []
-
-            for name in skills_before:
-                if name in consolidated:
-                    target = consolidated[name]
-                    mapped[name] = target
-                    if target and target not in new_skills:
-                        new_skills.append(target)
-                elif name in pruned_set:
-                    dropped.append(name)
-                else:
-                    if name not in new_skills:
-                        new_skills.append(name)
-
-            if not mapped and not dropped:
-                continue
-
-            job["skills"] = new_skills
-            job["skill"] = new_skills[0] if new_skills else None
-            changed = True
-
-            rewrites.append({
-                "job_id": job.get("id"),
-                "job_name": job.get("name") or job.get("id"),
-                "before": list(skills_before),
-                "after": list(new_skills),
-                "mapped": mapped,
-                "dropped": dropped,
-            })
-
-        if changed:
-            save_jobs(jobs)
-            logger.info(
-                "Curator rewrote skill references in %d cron job(s)", len(rewrites)
-            )
-
-        return {
-            "rewrites": rewrites,
-            "jobs_updated": len(rewrites),
-            "jobs_scanned": len(jobs),
-        }

docker-compose.yml

@@ -40,7 +40,7 @@ services:
       # - TEAMS_CLIENT_SECRET=${TEAMS_CLIENT_SECRET}
       # - TEAMS_TENANT_ID=${TEAMS_TENANT_ID}
       # - TEAMS_ALLOWED_USERS=${TEAMS_ALLOWED_USERS}
-      # - TEAMS_PORT=${TEAMS_PORT:-3978}
+      # - TEAMS_PORT=3978
     command: ["gateway", "run"]
 
   dashboard:

gateway/config.py

@@ -36,26 +36,6 @@ def _coerce_bool(value: Any, default: bool = True) -> bool:
     return is_truthy_value(value, default=default)
 
 
-def _coerce_float(value: Any, default: float) -> float:
-    """Coerce numeric config values, falling back on malformed input."""
-    if value is None:
-        return default
-    try:
-        return float(value)
-    except (TypeError, ValueError):
-        return default
-
-
-def _coerce_int(value: Any, default: int) -> int:
-    """Coerce integer config values, falling back on malformed input."""
-    if value is None:
-        return default
-    try:
-        return int(value)
-    except (TypeError, ValueError):
-        return default
-
-
 def _normalize_unauthorized_dm_behavior(value: Any, default: str = "pair") -> str:
     """Normalize unauthorized DM behavior to a supported value."""
     if isinstance(value, str):
@@ -321,13 +301,13 @@ class StreamingConfig:
         if not data:
             return cls()
         return cls(
-            enabled=_coerce_bool(data.get("enabled"), False),
+            enabled=data.get("enabled", False),
             transport=data.get("transport", "edit"),
-            edit_interval=_coerce_float(data.get("edit_interval"), 1.0),
-            buffer_threshold=_coerce_int(data.get("buffer_threshold"), 40),
+            edit_interval=float(data.get("edit_interval", 1.0)),
+            buffer_threshold=int(data.get("buffer_threshold", 40)),
             cursor=data.get("cursor", " ▉"),
-            fresh_final_after_seconds=_coerce_float(
-                data.get("fresh_final_after_seconds"), 60.0
+            fresh_final_after_seconds=float(
+                data.get("fresh_final_after_seconds", 60.0)
             ),
         )
 
@@ -920,12 +900,6 @@ def load_gateway_config() -> GatewayConfig:
                 if "dm_mention_threads" in matrix_cfg and not os.getenv("MATRIX_DM_MENTION_THREADS"):
                     os.environ["MATRIX_DM_MENTION_THREADS"] = str(matrix_cfg["dm_mention_threads"]).lower()
 
-            # Feishu settings → env vars (env vars take precedence)
-            feishu_cfg = yaml_cfg.get("feishu", {})
-            if isinstance(feishu_cfg, dict):
-                if "allow_bots" in feishu_cfg and not os.getenv("FEISHU_ALLOW_BOTS"):
-                    os.environ["FEISHU_ALLOW_BOTS"] = str(feishu_cfg["allow_bots"]).lower()
-
     except Exception as e:
         logger.warning(
             "Failed to process config.yaml — falling back to .env / gateway.json values. "
@@ -1077,14 +1051,7 @@ def _apply_env_overrides(config: GatewayConfig) -> None:
         if Platform.WHATSAPP not in config.platforms:
             config.platforms[Platform.WHATSAPP] = PlatformConfig()
         config.platforms[Platform.WHATSAPP].enabled = True
-    whatsapp_home = os.getenv("WHATSAPP_HOME_CHANNEL")
-    if whatsapp_home and Platform.WHATSAPP in config.platforms:
-        config.platforms[Platform.WHATSAPP].home_channel = HomeChannel(
-            platform=Platform.WHATSAPP,
-            chat_id=whatsapp_home,
-            name=os.getenv("WHATSAPP_HOME_CHANNEL_NAME", "Home"),
-        )
-
+    
     # Slack
     slack_token = os.getenv("SLACK_BOT_TOKEN")
     if slack_token:

gateway/platforms/api_server.py

@@ -2351,11 +2351,10 @@ class APIServerAdapter(BasePlatformAdapter):
             )
             if agent_ref is not None:
                 agent_ref[0] = agent
-            effective_task_id = session_id or str(uuid.uuid4())
             result = agent.run_conversation(
                 user_message=user_message,
                 conversation_history=conversation_history,
-                task_id=effective_task_id,
+                task_id="default",
             )
             usage = {
                 "input_tokens": getattr(agent, "session_prompt_tokens", 0) or 0,
@@ -2552,11 +2551,10 @@ class APIServerAdapter(BasePlatformAdapter):
                 )
                 self._active_run_agents[run_id] = agent
                 def _run_sync():
-                    effective_task_id = session_id or run_id
                     r = agent.run_conversation(
                         user_message=user_message,
                         conversation_history=conversation_history,
-                        task_id=effective_task_id,
+                        task_id="default",
                     )
                     u = {
                         "input_tokens": getattr(agent, "session_prompt_tokens", 0) or 0,

gateway/platforms/base.py

@@ -416,7 +416,7 @@ def is_host_excluded_by_no_proxy(hostname: str, no_proxy_value: str | None = Non
 from dataclasses import dataclass, field
 from datetime import datetime
 from pathlib import Path
-from typing import Dict, List, Optional, Any, Callable, Awaitable, Tuple, Union
+from typing import Dict, List, Optional, Any, Callable, Awaitable, Tuple
 from enum import Enum
 
 from pathlib import Path as _Path
@@ -981,7 +981,7 @@ def coerce_plaintext_gateway_command(event: "MessageEvent") -> None:
         return
 
 
-@dataclass
+@dataclass 
 class SendResult:
     """Result of sending a message."""
     success: bool
@@ -991,45 +991,6 @@ class SendResult:
     retryable: bool = False  # True for transient connection errors — base will retry automatically
 
 
-class EphemeralReply(str):
-    """System-notice reply that auto-deletes after a TTL.
-
-    Slash-command handlers in ``gateway/run.py`` can return this wrapper
-    instead of a plain string to request that the reply message be deleted
-    after ``ttl_seconds`` on platforms that support ``delete_message``.
-
-    Subclassing ``str`` keeps the wrapper transparent to anything that
-    treats handler return values as text (existing tests use ``in`` /
-    ``startswith`` / equality; the ``_process_message_background`` pipeline
-    extracts attachments from the string content).  ``isinstance(r,
-    EphemeralReply)`` still distinguishes ephemeral replies from plain
-    strings so the send path can schedule deletion.
-
-    Platforms that don't override :meth:`BasePlatformAdapter.delete_message`
-    silently ignore the TTL — the message is sent normally and left in
-    place.  When ``ttl_seconds`` is ``None``, the pipeline uses the
-    configured ``display.ephemeral_system_ttl`` default.  A default of ``0``
-    disables auto-deletion globally, preserving prior behavior.
-    """
-
-    ttl_seconds: Optional[int]
-
-    def __new__(cls, text: str, ttl_seconds: Optional[int] = None):
-        instance = super().__new__(cls, text)
-        instance.ttl_seconds = ttl_seconds
-        return instance
-
-    @property
-    def text(self) -> str:
-        """Return the underlying text.
-
-        Provided for call sites that want an explicit string conversion,
-        though ``str(reply)`` and using ``reply`` directly where a string
-        is expected both work identically.
-        """
-        return str.__str__(self)
-
-
 def merge_pending_message_event(
     pending_messages: Dict[str, MessageEvent],
     session_key: str,
@@ -1073,11 +1034,6 @@ def merge_pending_message_event(
                     existing.text = event.text
             if existing_is_photo or incoming_is_photo:
                 existing.message_type = MessageType.PHOTO
-            elif (
-                getattr(existing, "message_type", None) == MessageType.TEXT
-                and event.message_type != MessageType.TEXT
-            ):
-                existing.message_type = event.message_type
             return
 
         if (
@@ -1112,10 +1068,8 @@ _RETRYABLE_ERROR_PATTERNS = (
 )
 
 
-# Type for message handlers.  Handlers may return a plain string (normal
-# reply), an ``EphemeralReply`` to opt the reply into auto-deletion, or
-# ``None`` when the response was already delivered (e.g. via streaming).
-MessageHandler = Callable[[MessageEvent], Awaitable[Optional[Union[str, "EphemeralReply"]]]]
+# Type for message handlers
+MessageHandler = Callable[[MessageEvent], Awaitable[Optional[str]]]
 
 
 def resolve_channel_prompt(
@@ -1500,64 +1454,6 @@ class BasePlatformAdapter(ABC):
         """
         return False
 
-    def _get_ephemeral_system_ttl_default(self) -> int:
-        """Read ``display.ephemeral_system_ttl`` from config.
-
-        Returns the TTL in seconds to use when an :class:`EphemeralReply`
-        does not specify one explicitly.  ``0`` (the default) disables
-        auto-deletion.  Non-fatal if config is unreadable.
-        """
-        try:
-            from hermes_cli.config import load_config as _load_config
-        except Exception:
-            return 0
-        try:
-            cfg = _load_config()
-        except Exception:
-            return 0
-        display = cfg.get("display", {}) if isinstance(cfg, dict) else {}
-        if not isinstance(display, dict):
-            return 0
-        raw = display.get("ephemeral_system_ttl", 0)
-        try:
-            return int(raw)
-        except (TypeError, ValueError):
-            return 0
-
-    def _schedule_ephemeral_delete(
-        self,
-        chat_id: str,
-        message_id: str,
-        ttl_seconds: int,
-    ) -> None:
-        """Spawn a detached task that deletes ``message_id`` after ``ttl_seconds``.
-
-        Best-effort — failures (gateway restart, permission denied, message
-        too old for Telegram's 48h window) are swallowed at debug level.
-        Does not block the caller.
-        """
-
-        async def _run_delete() -> None:
-            try:
-                await asyncio.sleep(max(1, int(ttl_seconds)))
-                await self.delete_message(chat_id=chat_id, message_id=message_id)
-            except asyncio.CancelledError:
-                raise
-            except Exception as e:
-                logger.debug(
-                    "[%s] Ephemeral delete failed for %s/%s: %s",
-                    self.name, chat_id, message_id, e,
-                )
-
-        coro = _run_delete()
-        try:
-            asyncio.create_task(coro)
-        except RuntimeError:
-            # No running loop (e.g. unit tests that never reach the async
-            # path).  Close the coroutine cleanly so Python doesn't warn
-            # about it never being awaited, then drop silently.
-            coro.close()
-
     async def send_slash_confirm(
         self,
         chat_id: str,
@@ -2147,28 +2043,6 @@ class BasePlatformAdapter(ABC):
         lowered = error.lower()
         return "timed out" in lowered or "readtimeout" in lowered or "writetimeout" in lowered
 
-    def _unwrap_ephemeral(self, response: Any) -> Tuple[Optional[str], int]:
-        """Unwrap a handler response into (text, ttl_seconds).
-
-        Accepts a plain string, ``None``, or an :class:`EphemeralReply`.
-        Returns ``(text, ttl)`` where ``ttl > 0`` means the caller should
-        schedule a deletion via :meth:`_schedule_ephemeral_delete` after
-        the send succeeds.  ``ttl`` is forced to 0 when the adapter
-        doesn't override :meth:`delete_message` so non-supporting
-        platforms silently degrade to normal sends.
-        """
-        if isinstance(response, EphemeralReply):
-            ttl = response.ttl_seconds
-            if ttl is None:
-                try:
-                    ttl = int(self._get_ephemeral_system_ttl_default())
-                except Exception:
-                    ttl = 0
-            if ttl and ttl > 0 and type(self).delete_message is BasePlatformAdapter.delete_message:
-                ttl = 0
-            return response.text, int(ttl or 0)
-        return response, 0
-
     async def _send_with_retry(
         self,
         chat_id: str,
@@ -2476,20 +2350,13 @@ class BasePlatformAdapter(ABC):
                 release_guard=False,
                 discard_pending=False,
             )
-            _text, _eph_ttl = self._unwrap_ephemeral(response)
-            if _text:
-                _r = await self._send_with_retry(
+            if response:
+                await self._send_with_retry(
                     chat_id=event.source.chat_id,
-                    content=_text,
+                    content=response,
                     reply_to=event.message_id,
                     metadata=thread_meta,
                 )
-                if _eph_ttl > 0 and _r.success and _r.message_id:
-                    self._schedule_ephemeral_delete(
-                        chat_id=event.source.chat_id,
-                        message_id=_r.message_id,
-                        ttl_seconds=_eph_ttl,
-                    )
         except Exception:
             # On failure, restore the original guard if one still exists so
             # we don't leave the session in a half-reset state.
@@ -2569,20 +2436,13 @@ class BasePlatformAdapter(ABC):
                 try:
                     _thread_meta = {"thread_id": event.source.thread_id} if event.source.thread_id else None
                     response = await self._message_handler(event)
-                    _text, _eph_ttl = self._unwrap_ephemeral(response)
-                    if _text:
-                        _r = await self._send_with_retry(
+                    if response:
+                        await self._send_with_retry(
                             chat_id=event.source.chat_id,
-                            content=_text,
+                            content=response,
                             reply_to=event.message_id,
                             metadata=_thread_meta,
                         )
-                        if _eph_ttl > 0 and _r.success and _r.message_id:
-                            self._schedule_ephemeral_delete(
-                                chat_id=event.source.chat_id,
-                                message_id=_r.message_id,
-                                ttl_seconds=_eph_ttl,
-                            )
                 except Exception as e:
                     logger.error("[%s] Command '/%s' dispatch failed: %s", self.name, cmd, e, exc_info=True)
                 return
@@ -2656,6 +2516,7 @@ class BasePlatformAdapter(ABC):
         # Fall back to a new Event only if the entry was removed externally.
         interrupt_event = self._active_sessions.get(session_key) or asyncio.Event()
         self._active_sessions[session_key] = interrupt_event
+        callback_generation = getattr(interrupt_event, "_hermes_run_generation", None)
         
         # Start continuous typing indicator (refreshes every 2 seconds)
         _thread_metadata = {"thread_id": event.source.thread_id} if event.source.thread_id else None
@@ -2688,16 +2549,7 @@ class BasePlatformAdapter(ABC):
 
             # Call the handler (this can take a while with tool calls)
             response = await self._message_handler(event)
-
-            # Slash-command handlers may return an EphemeralReply sentinel to
-            # request that their reply message auto-delete after a TTL (used
-            # for system notices like "✨ New session started!" that the user
-            # doesn't need to keep in the thread).  Unwrap here so all the
-            # downstream extract_media / text-processing logic sees a plain
-            # string, and remember the TTL + platform capability so the
-            # post-send block can schedule the deletion.
-            response, _ephemeral_ttl = self._unwrap_ephemeral(response)
-
+            
             # Send response if any.  A None/empty response is normal when
             # streaming already delivered the text (already_sent=True) or
             # when the message was queued behind an active agent.  Log at
@@ -2786,21 +2638,6 @@ class BasePlatformAdapter(ABC):
                     )
                     _record_delivery(result)
 
-                    # Schedule auto-deletion of system-notice replies.
-                    # Detached so the handler returns immediately; errors
-                    # (permission denied, message too old) are swallowed.
-                    if (
-                        _ephemeral_ttl
-                        and _ephemeral_ttl > 0
-                        and result.success
-                        and result.message_id
-                    ):
-                        self._schedule_ephemeral_delete(
-                            chat_id=event.source.chat_id,
-                            message_id=result.message_id,
-                            ttl_seconds=_ephemeral_ttl,
-                        )
-
                 # Human-like pacing delay between text and media
                 human_delay = self._get_human_delay()
 
@@ -2978,20 +2815,7 @@ class BasePlatformAdapter(ABC):
         finally:
             # Fire any one-shot post-delivery callback registered for this
             # session (e.g. deferred background-review notifications).
-            #
-            # Snapshot the callback generation HERE (after the agent has run),
-            # not at the top of this task.  _hermes_run_generation is set on
-            # the interrupt event by GatewayRunner._bind_adapter_run_generation
-            # during _handle_message_with_agent — which happens DURING the
-            # self._message_handler(event) await above.  Snapshotting earlier
-            # always captured None, which bypassed the generation-ownership
-            # check in pop_post_delivery_callback and let stale runs fire a
-            # fresher run's callbacks.
-            _callback_generation = getattr(
-                interrupt_event,
-                "_hermes_run_generation",
-                None,
-            )
+            _callback_generation = callback_generation
             if hasattr(self, "pop_post_delivery_callback"):
                 _post_cb = self.pop_post_delivery_callback(
                     session_key,

gateway/platforms/discord.py

@@ -3078,7 +3078,6 @@ class DiscordAdapter(BasePlatformAdapter):
     async def send_update_prompt(
         self, chat_id: str, prompt: str, default: str = "",
         session_key: str = "",
-        metadata: Optional[Dict[str, Any]] = None,
     ) -> SendResult:
         """Send an interactive button-based update prompt (Yes / No).
 
@@ -3088,10 +3087,9 @@ class DiscordAdapter(BasePlatformAdapter):
         if not self._client or not DISCORD_AVAILABLE:
             return SendResult(success=False, error="Not connected")
         try:
-            target_id = metadata.get("thread_id") if metadata and metadata.get("thread_id") else chat_id
-            channel = self._client.get_channel(int(target_id))
+            channel = self._client.get_channel(int(chat_id))
             if not channel:
-                channel = await self._client.fetch_channel(int(target_id))
+                channel = await self._client.fetch_channel(int(chat_id))
 
             default_hint = f" (default: {default})" if default else ""
             embed = discord.Embed(

gateway/platforms/feishu.py

@@ -64,7 +64,7 @@ from dataclasses import dataclass, field
 from datetime import datetime
 from pathlib import Path
 from types import SimpleNamespace
-from typing import Any, Dict, List, Literal, Optional, Sequence
+from typing import Any, Dict, List, Optional, Sequence
 from urllib.error import HTTPError, URLError
 from urllib.parse import urlencode
 from urllib.request import Request, urlopen
@@ -141,7 +141,6 @@ from gateway.platforms.base import (
 )
 from gateway.status import acquire_scoped_lock, release_scoped_lock
 from hermes_constants import get_hermes_home
-from utils import atomic_json_write
 
 logger = logging.getLogger(__name__)
 
@@ -388,8 +387,6 @@ class FeishuAdapterSettings:
     admins: frozenset[str] = frozenset()
     default_group_policy: str = ""
     group_rules: Dict[str, FeishuGroupRule] = field(default_factory=dict)
-    allow_bots: str = "none"  # "none" | "mentions" | "all"
-    require_mention: bool = True
 
 
 @dataclass
@@ -399,7 +396,6 @@ class FeishuGroupRule:
     policy: str  # "open" | "allowlist" | "blacklist" | "admin_only" | "disabled"
     allowlist: set[str] = field(default_factory=set)
     blacklist: set[str] = field(default_factory=set)
-    require_mention: Optional[bool] = None  # None = inherit global
 
 
 @dataclass
@@ -409,40 +405,6 @@ class FeishuBatchState:
     counts: Dict[str, int] = field(default_factory=dict)
 
 
-# ---------------------------------------------------------------------------
-# Admission: policy types
-# ---------------------------------------------------------------------------
-
-
-RejectReason = Literal[
-    "self_echo",
-    "self_ids_unknown",
-    "bots_disabled",
-    "bot_not_mentioned",
-    "group_policy_rejected",
-]
-
-
-def _is_bot_sender(sender: Any) -> bool:
-    # receive_v1 docs say {user, bot}; accept "app" defensively.
-    return getattr(sender, "sender_type", "") in ("bot", "app")
-
-
-def _sender_identity(sender: Any) -> frozenset:
-    # Take any non-empty id variant — tenant sender_id_type decides which are populated.
-    sid = getattr(sender, "sender_id", None)
-    if sid is None:
-        return frozenset()
-    return frozenset(
-        v for v in (
-            getattr(sid, "open_id", None),
-            getattr(sid, "user_id", None),
-            getattr(sid, "union_id", None),
-        )
-        if v
-    )
-
-
 # ---------------------------------------------------------------------------
 # Markdown rendering helpers
 # ---------------------------------------------------------------------------
@@ -1415,16 +1377,10 @@ class FeishuAdapter(BasePlatformAdapter):
             for chat_id, rule_cfg in raw_group_rules.items():
                 if not isinstance(rule_cfg, dict):
                     continue
-                # Only override when the key is explicitly set — missing vs false
-                # must not collapse.
-                per_chat_require_mention: Optional[bool] = None
-                if "require_mention" in rule_cfg:
-                    per_chat_require_mention = _to_boolean(rule_cfg.get("require_mention"))
                 group_rules[str(chat_id)] = FeishuGroupRule(
                     policy=str(rule_cfg.get("policy", "open")).strip().lower(),
                     allowlist=set(str(u).strip() for u in rule_cfg.get("allowlist", []) if str(u).strip()),
                     blacklist=set(str(u).strip() for u in rule_cfg.get("blacklist", []) if str(u).strip()),
-                    require_mention=per_chat_require_mention,
                 )
 
         # Bot-level admins
@@ -1434,16 +1390,6 @@ class FeishuAdapter(BasePlatformAdapter):
         # Default group policy (for groups not in group_rules)
         default_group_policy = str(extra.get("default_group_policy", "")).strip().lower()
 
-        # Env-only so adapter and gateway auth bypass share one source; yaml
-        # feishu.allow_bots is bridged to this env var at config load.
-        allow_bots = os.getenv("FEISHU_ALLOW_BOTS", "none").strip().lower()
-        if allow_bots not in ("none", "mentions", "all"):
-            logger.warning(
-                "[Feishu] Unknown allow_bots=%r, falling back to 'none'. Valid: none, mentions, all.",
-                allow_bots,
-            )
-            allow_bots = "none"
-
         return FeishuAdapterSettings(
             app_id=str(extra.get("app_id") or os.getenv("FEISHU_APP_ID", "")).strip(),
             app_secret=str(extra.get("app_secret") or os.getenv("FEISHU_APP_SECRET", "")).strip(),
@@ -1500,10 +1446,6 @@ class FeishuAdapter(BasePlatformAdapter):
             admins=admins,
             default_group_policy=default_group_policy,
             group_rules=group_rules,
-            allow_bots=allow_bots,
-            require_mention=_to_boolean(
-                extra.get("require_mention", os.getenv("FEISHU_REQUIRE_MENTION", "true"))
-            ),
         )
 
     def _apply_settings(self, settings: FeishuAdapterSettings) -> None:
@@ -1534,8 +1476,6 @@ class FeishuAdapter(BasePlatformAdapter):
         self._ws_reconnect_interval = settings.ws_reconnect_interval
         self._ws_ping_interval = settings.ws_ping_interval
         self._ws_ping_timeout = settings.ws_ping_timeout
-        self._allow_bots = settings.allow_bots
-        self._require_mention = settings.require_mention
 
     def _build_event_handler(self) -> Any:
         if EventDispatcherHandler is None:
@@ -2249,28 +2189,30 @@ class FeishuAdapter(BasePlatformAdapter):
         event = getattr(data, "event", None)
         message = getattr(event, "message", None)
         sender = getattr(event, "sender", None)
-        if not message or not sender or not getattr(sender, "sender_id", None):
-            logger.debug("[Feishu] Dropping malformed inbound event: missing message/sender")
+        sender_id = getattr(sender, "sender_id", None)
+        if not message or not sender_id:
+            logger.debug("[Feishu] Dropping malformed inbound event: missing message or sender_id")
             return
 
         message_id = getattr(message, "message_id", None)
         if not message_id or self._is_duplicate(message_id):
             logger.debug("[Feishu] Dropping duplicate/missing message_id: %s", message_id)
             return
-
-        reason = self._admit(sender, message)
-        if reason is not None:
-            logger.debug("[Feishu] dropping inbound event: %s", reason)
+        if self._is_self_sent_bot_message(event):
+            logger.debug("[Feishu] Dropping self-sent bot event: %s", message_id)
             return
 
         chat_type = getattr(message, "chat_type", "p2p")
+        chat_id = getattr(message, "chat_id", "") or ""
+        if chat_type != "p2p" and not self._should_accept_group_message(message, sender_id, chat_id):
+            logger.debug("[Feishu] Dropping group message that failed mention/policy gate: %s", message_id)
+            return
         await self._process_inbound_message(
             data=data,
             message=message,
-            sender_id=getattr(sender, "sender_id", None),
+            sender_id=sender_id,
             chat_type=chat_type,
             message_id=message_id,
-            is_bot=_is_bot_sender(sender),
         )
 
     def _on_message_read_event(self, data: P2ImMessageMessageReadV1) -> None:
@@ -2447,11 +2389,10 @@ class FeishuAdapter(BasePlatformAdapter):
             msg = items[0] if items else None
             if not msg:
                 return
-            # GET im/v1/messages returns sender.id=app_id for bot messages —
-            # peer bots and us share sender_type="app" but differ on app_id.
             sender = getattr(msg, "sender", None)
-            if str(getattr(sender, "id", "") or "") != self._app_id:
-                return  # only route reactions on this bot's own messages
+            sender_type = str(getattr(sender, "sender_type", "") or "").lower()
+            if sender_type != "app":
+                return  # only route reactions on our own bot messages
             chat_id = str(getattr(msg, "chat_id", "") or "")
             chat_type_raw = str(getattr(msg, "chat_type", "p2p") or "p2p")
             if not chat_id:
@@ -2738,7 +2679,6 @@ class FeishuAdapter(BasePlatformAdapter):
         sender_id: Any,
         chat_type: str,
         message_id: str,
-        is_bot: bool = False,
     ) -> None:
         text, inbound_type, media_urls, media_types, mentions = await self._extract_message_content(message)
 
@@ -2764,27 +2704,19 @@ class FeishuAdapter(BasePlatformAdapter):
         )
         reply_to_text = await self._fetch_message_text(reply_to_message_id) if reply_to_message_id else None
 
-        sender_primary = (
-            getattr(sender_id, "open_id", None)
-            or getattr(sender_id, "user_id", None)
-            or getattr(sender_id, "union_id", None)
-            or "<unknown>"
-        )
         logger.info(
-            "[Feishu] Inbound %s message received: id=%s type=%s chat_id=%s sender=%s:%s text=%r media=%d",
+            "[Feishu] Inbound %s message received: id=%s type=%s chat_id=%s text=%r media=%d",
             "dm" if chat_type == "p2p" else "group",
             message_id,
             inbound_type.value,
             getattr(message, "chat_id", "") or "",
-            "bot" if is_bot else "user",
-            sender_primary,
             text[:120],
             len(media_urls),
         )
 
         chat_id = getattr(message, "chat_id", "") or ""
         chat_info = await self.get_chat_info(chat_id)
-        sender_profile = await self._resolve_sender_profile(sender_id, is_bot=is_bot)
+        sender_profile = await self._resolve_sender_profile(sender_id)
         source = self.build_source(
             chat_id=chat_id,
             chat_name=chat_info.get("name") or chat_id or "Feishu Chat",
@@ -2793,7 +2725,6 @@ class FeishuAdapter(BasePlatformAdapter):
             user_name=sender_profile["user_name"],
             thread_id=getattr(message, "thread_id", None) or None,
             user_id_alt=sender_profile["user_id_alt"],
-            is_bot=is_bot,
         )
         normalized = MessageEvent(
             text=text,
@@ -3516,12 +3447,7 @@ class FeishuAdapter(BasePlatformAdapter):
             return "dm"
         return "group"
 
-    async def _resolve_sender_profile(
-        self,
-        sender_id: Any,
-        *,
-        is_bot: bool = False,
-    ) -> Dict[str, Optional[str]]:
+    async def _resolve_sender_profile(self, sender_id: Any) -> Dict[str, Optional[str]]:
         """Map Feishu's three-tier user IDs onto Hermes' SessionSource fields.
 
         Preference order for the primary ``user_id`` field:
@@ -3538,11 +3464,7 @@ class FeishuAdapter(BasePlatformAdapter):
         union_id = getattr(sender_id, "union_id", None) or None
         # Prefer tenant-scoped user_id; fall back to app-scoped open_id.
         primary_id = user_id or open_id
-        # bot/v3/bots/basic_batch only accepts open_id.
-        name_lookup_id = open_id if is_bot else (primary_id or union_id)
-        display_name = await self._resolve_sender_name_from_api(
-            name_lookup_id, is_bot=is_bot,
-        )
+        display_name = await self._resolve_sender_name_from_api(primary_id or union_id)
         return {
             "user_id": primary_id,
             "user_name": display_name,
@@ -3562,14 +3484,11 @@ class FeishuAdapter(BasePlatformAdapter):
         self._sender_name_cache.pop(sender_id, None)
         return None
 
-    async def _resolve_sender_name_from_api(
-        self,
-        sender_id: Optional[str],
-        *,
-        is_bot: bool = False,
-    ) -> Optional[str]:
-        """Bots divert to bot/basic_batch — contact API doesn't return bot names.
-        Failures are silent so the pipeline never blocks on name resolution.
+    async def _resolve_sender_name_from_api(self, sender_id: Optional[str]) -> Optional[str]:
+        """Fetch the sender's display name from the Feishu contact API with a 10-minute cache.
+
+        ID-type detection mirrors openclaw: ou_ → open_id, on_ → union_id, else user_id.
+        Failures are silently suppressed; the message pipeline must not block on name resolution.
         """
         if not sender_id or not self._client:
             return None
@@ -3579,16 +3498,7 @@ class FeishuAdapter(BasePlatformAdapter):
         now = time.time()
         cached_name = self._get_cached_sender_name(trimmed)
         if cached_name is not None:
-            return cached_name or None  # "" cached means "known nameless"
-        if is_bot:
-            names = await self._fetch_bot_names([trimmed])
-            if names is None:
-                return None
-            expire_at = now + _FEISHU_SENDER_NAME_TTL_SECONDS
-            for oid, name in names.items():
-                self._sender_name_cache[oid] = (name, expire_at)
-            hit = self._sender_name_cache.get(trimmed)
-            return (hit[0] or None) if hit else None
+            return cached_name
         try:
             from lark_oapi.api.contact.v3 import GetUserRequest  # lazy import
             if trimmed.startswith("ou_"):
@@ -3617,35 +3527,6 @@ class FeishuAdapter(BasePlatformAdapter):
             logger.debug("[Feishu] Failed to resolve sender name for %s", sender_id, exc_info=True)
         return None
 
-    async def _fetch_bot_names(self, bot_ids: List[str]) -> Optional[Dict[str, str]]:
-        if not self._client or not bot_ids:
-            return None
-        try:
-            req = (
-                BaseRequest.builder()
-                .http_method(HttpMethod.GET)
-                .uri("/open-apis/bot/v3/bots/basic_batch")
-                .queries([("bot_ids", oid) for oid in bot_ids])
-                .token_types({AccessTokenType.TENANT})
-                .build()
-            )
-            resp = await asyncio.to_thread(self._client.request, req)
-            content = getattr(getattr(resp, "raw", None), "content", None)
-            if not content:
-                return None
-            payload = json.loads(content)
-            if payload.get("code") != 0:
-                return None
-            bots = (payload.get("data") or {}).get("bots") or {}
-            return {
-                oid: str(info.get("name") or "").strip()
-                for oid, info in bots.items()
-                if oid
-            }
-        except Exception:
-            logger.debug("[Feishu] Failed to fetch bot names for %s", bot_ids, exc_info=True)
-            return None
-
     async def _fetch_message_text(self, message_id: str) -> Optional[str]:
         if not self._client or not message_id:
             return None
@@ -3709,60 +3590,10 @@ class FeishuAdapter(BasePlatformAdapter):
             logger.exception("[Feishu] Background inbound processing failed")
 
     # =========================================================================
-    # Inbound admission
+    # Group policy and mention gating
     # =========================================================================
 
-    def _admit(self, sender: Any, message: Any) -> Optional[RejectReason]:
-        sender_ids = _sender_identity(sender)
-        self_ids = frozenset(v for v in (self._bot_open_id, self._bot_user_id) if v)
-        is_bot = _is_bot_sender(sender)
-        is_group = getattr(message, "chat_type", "p2p") != "p2p"
-        chat_id = getattr(message, "chat_id", "") or ""
-        require_mention = is_group and self._require_mention_for(chat_id)
-
-        # Defensive only — Feishu doesn't echo our outbound back as inbound,
-        # and open_id is always populated on both sides.
-        if self_ids and sender_ids & self_ids:
-            return "self_echo"
-
-        if is_bot:
-            mode = self._allow_bots
-            if mode != "mentions" and mode != "all":
-                return "bots_disabled"
-            # Defensive: pre-hydration or malformed payloads.
-            if not self_ids or not sender_ids:
-                return "self_ids_unknown"
-            # Step 4 covers mention enforcement for groups when require_mention
-            # is on; check here only on paths step 4 won't reach.
-            if mode == "mentions" and not require_mention and not self._mentions_self(message):
-                return "bot_not_mentioned"
-
-        if not is_group:
-            return None
-
-        if not self._allow_group_message(
-            getattr(sender, "sender_id", None), chat_id, is_bot=is_bot,
-        ):
-            return "group_policy_rejected"
-        if require_mention and not self._mentions_self(message):
-            return "group_policy_rejected"
-        return None
-
-    def _require_mention_for(self, chat_id: str) -> bool:
-        rule = self._group_rules.get(chat_id) if chat_id else None
-        if rule and rule.require_mention is not None:
-            return rule.require_mention
-        return self._require_mention
-
-    # --- Group policy ---------------------------------------------------------
-
-    def _allow_group_message(
-        self,
-        sender_id: Any,
-        chat_id: str = "",
-        *,
-        is_bot: bool = False,
-    ) -> bool:
+    def _allow_group_message(self, sender_id: Any, chat_id: str = "") -> bool:
         """Per-group policy gate for non-DM traffic."""
         sender_open_id = getattr(sender_id, "open_id", None)
         sender_user_id = getattr(sender_id, "user_id", None)
@@ -3781,17 +3612,12 @@ class FeishuAdapter(BasePlatformAdapter):
             allowlist = self._allowed_group_users
             blacklist = set()
 
-        # Channel locks apply to everyone; allowlist/blacklist only gate humans
-        # (bots were already cleared upstream by FEISHU_ALLOW_BOTS).
         if policy == "disabled":
             return False
         if policy == "open":
             return True
         if policy == "admin_only":
             return False
-        if is_bot:
-            return True
-
         if policy == "allowlist":
             return bool(sender_ids and (sender_ids & allowlist))
         if policy == "blacklist":
@@ -3799,16 +3625,17 @@ class FeishuAdapter(BasePlatformAdapter):
 
         return bool(sender_ids and (sender_ids & self._allowed_group_users))
 
-    # --- Mention detection ----------------------------------------------------
-
-    def _mentions_self(self, message: Any) -> bool:
-        # @_all is Feishu's @everyone placeholder.
+    def _should_accept_group_message(self, message: Any, sender_id: Any, chat_id: str = "") -> bool:
+        """Require an explicit @mention before group messages enter the agent."""
+        if not self._allow_group_message(sender_id, chat_id):
+            return False
+        # @_all is Feishu's @everyone placeholder — always route to the bot.
         raw_content = getattr(message, "content", "") or ""
         if "@_all" in raw_content:
             return True
         mentions = getattr(message, "mentions", None) or []
-        if mentions and self._message_mentions_bot(mentions):
-            return True
+        if mentions:
+            return self._message_mentions_bot(mentions)
         normalized = normalize_feishu_message(
             message_type=getattr(message, "message_type", "") or "",
             raw_content=raw_content,
@@ -3817,6 +3644,23 @@ class FeishuAdapter(BasePlatformAdapter):
         )
         return self._post_mentions_bot(normalized.mentions)
 
+    def _is_self_sent_bot_message(self, event: Any) -> bool:
+        """Return True only for Feishu events emitted by this Hermes bot."""
+        sender = getattr(event, "sender", None)
+        sender_type = str(getattr(sender, "sender_type", "") or "").strip().lower()
+        if sender_type not in {"bot", "app"}:
+            return False
+
+        sender_id = getattr(sender, "sender_id", None)
+        sender_open_id = str(getattr(sender_id, "open_id", "") or "").strip()
+        sender_user_id = str(getattr(sender_id, "user_id", "") or "").strip()
+
+        if self._bot_open_id and sender_open_id == self._bot_open_id:
+            return True
+        if self._bot_user_id and sender_user_id == self._bot_user_id:
+            return True
+        return False
+
     def _message_mentions_bot(self, mentions: List[Any]) -> bool:
         # IDs trump names: when both sides have open_id (or both user_id),
         # match requires equal IDs. Name fallback only when either side
@@ -3960,7 +3804,7 @@ class FeishuAdapter(BasePlatformAdapter):
             recent = self._seen_message_order[-self._dedup_cache_size:]
             # Save as {msg_id: timestamp} so TTL filtering works across restarts.
             payload = {"message_ids": {k: self._seen_message_ids[k] for k in recent if k in self._seen_message_ids}}
-            atomic_json_write(self._dedup_state_path, payload, indent=None)
+            self._dedup_state_path.write_text(json.dumps(payload, ensure_ascii=False), encoding="utf-8")
         except OSError:
             logger.warning("[Feishu] Failed to persist dedup state to %s", self._dedup_state_path, exc_info=True)
 

gateway/platforms/helpers.py

@@ -13,8 +13,6 @@ import time
 from pathlib import Path
 from typing import TYPE_CHECKING, Dict
 
-from utils import atomic_json_write
-
 if TYPE_CHECKING:
     from gateway.platforms.base import MessageEvent
 
@@ -239,11 +237,12 @@ class ThreadParticipationTracker:
 
     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)
-        atomic_json_write(path, thread_list, indent=None)
+        path.write_text(json.dumps(thread_list), encoding="utf-8")
 
     def mark(self, thread_id: str) -> None:
         """Mark *thread_id* as participated and persist."""

gateway/platforms/signal.py

@@ -534,18 +534,6 @@ class SignalAdapter(BasePlatformAdapter):
                 except Exception:
                     logger.exception("Signal: failed to fetch attachment %s", att_id)
 
-        # Skip envelopes with no meaningful content (no text, no attachments).
-        # Catches profile key updates, empty messages, and other metadata-only
-        # envelopes that still carry a dataMessage wrapper but have nothing
-        # worth processing. See issue: signal-cli logs "Profile key update" +
-        # Hermes receives msg='' triggering a full agent turn for nothing.
-        if (not text or not text.strip()) and not media_urls:
-            logger.debug(
-                "Signal: skipping contentless envelope from %s (%d attachments)",
-                redact_phone(sender), len(media_urls) if media_urls else 0,
-            )
-            return
-
         # Build session source
         source = self.build_source(
             chat_id=chat_id,

gateway/platforms/telegram.py

@@ -290,53 +290,14 @@ class TelegramAdapter(BasePlatformAdapter):
         # and any other slash-confirm prompts; see GatewayRunner._request_slash_confirm).
         self._slash_confirm_state: Dict[str, str] = {}
 
-    def _is_callback_user_authorized(
-        self,
-        user_id: str,
-        *,
-        chat_id: Optional[str] = None,
-        chat_type: Optional[str] = None,
-        thread_id: Optional[str] = None,
-        user_name: Optional[str] = None,
-    ) -> bool:
+    @staticmethod
+    def _is_callback_user_authorized(user_id: str) -> bool:
         """Return whether a Telegram inline-button caller may perform gated actions."""
-        normalized_user_id = str(user_id or "").strip()
-        if not normalized_user_id:
-            return False
-
-        runner = getattr(getattr(self, "_message_handler", None), "__self__", None)
-        auth_fn = getattr(runner, "_is_user_authorized", None)
-        if callable(auth_fn):
-            try:
-                from gateway.session import SessionSource
-
-                normalized_chat_type = str(chat_type or "dm").strip().lower() or "dm"
-                if normalized_chat_type == "private":
-                    normalized_chat_type = "dm"
-                elif normalized_chat_type == "supergroup":
-                    normalized_chat_type = "forum" if thread_id is not None else "group"
-
-                source = SessionSource(
-                    platform=Platform.TELEGRAM,
-                    chat_id=str(chat_id or normalized_user_id),
-                    chat_type=normalized_chat_type,
-                    user_id=normalized_user_id,
-                    user_name=str(user_name).strip() if user_name else None,
-                    thread_id=str(thread_id) if thread_id is not None else None,
-                )
-                return bool(auth_fn(source))
-            except Exception:
-                logger.debug(
-                    "[Telegram] Falling back to env-only callback auth for user %s",
-                    normalized_user_id,
-                    exc_info=True,
-                )
-
         allowed_csv = os.getenv("TELEGRAM_ALLOWED_USERS", "").strip()
         if not allowed_csv:
             return True
         allowed_ids = {uid.strip() for uid in allowed_csv.split(",") if uid.strip()}
-        return "*" in allowed_ids or normalized_user_id in allowed_ids
+        return "*" in allowed_ids or user_id in allowed_ids
 
     @classmethod
     def _metadata_thread_id(cls, metadata: Optional[Dict[str, Any]]) -> Optional[str]:
@@ -1360,7 +1321,6 @@ class TelegramAdapter(BasePlatformAdapter):
     async def send_update_prompt(
         self, chat_id: str, prompt: str, default: str = "",
         session_key: str = "",
-        metadata: Optional[Dict[str, Any]] = None,
     ) -> SendResult:
         """Send an inline-keyboard update prompt (Yes / No buttons).
 
@@ -1378,14 +1338,11 @@ class TelegramAdapter(BasePlatformAdapter):
                     InlineKeyboardButton("✗ No", callback_data="update_prompt:n"),
                 ]
             ])
-            thread_id = self._metadata_thread_id(metadata)
-            message_thread_id = self._message_thread_id_for_send(thread_id)
             msg = await self._bot.send_message(
                 chat_id=int(chat_id),
                 text=text,
                 parse_mode=ParseMode.MARKDOWN,
                 reply_markup=keyboard,
-                message_thread_id=message_thread_id,
                 **self._link_preview_kwargs(),
             )
             return SendResult(success=True, message_id=str(msg.message_id))
@@ -1803,12 +1760,6 @@ class TelegramAdapter(BasePlatformAdapter):
         if not query or not query.data:
             return
         data = query.data
-        query_message = getattr(query, "message", None)
-        query_chat_id = getattr(query_message, "chat_id", None)
-        query_chat = getattr(query_message, "chat", None)
-        query_chat_type = getattr(query_chat, "type", None)
-        query_thread_id = getattr(query_message, "message_thread_id", None)
-        query_user_name = getattr(query.from_user, "first_name", None)
 
         # --- Model picker callbacks ---
         if data.startswith(("mp:", "mm:", "mb", "mx", "mg:")):
@@ -1830,13 +1781,7 @@ class TelegramAdapter(BasePlatformAdapter):
 
                 # Only authorized users may click approval buttons.
                 caller_id = str(getattr(query.from_user, "id", ""))
-                if not self._is_callback_user_authorized(
-                    caller_id,
-                    chat_id=query_chat_id,
-                    chat_type=str(query_chat_type) if query_chat_type is not None else None,
-                    thread_id=str(query_thread_id) if query_thread_id is not None else None,
-                    user_name=query_user_name,
-                ):
+                if not self._is_callback_user_authorized(caller_id):
                     await query.answer(text="⛔ You are not authorized to approve commands.")
                     return
 
@@ -1886,14 +1831,8 @@ class TelegramAdapter(BasePlatformAdapter):
                 choice = parts[1]  # once, always, cancel
                 confirm_id = parts[2]
 
-                caller_id = str(getattr(query.from_user, "id", ""))
-                if not self._is_callback_user_authorized(
-                    caller_id,
-                    chat_id=query_chat_id,
-                    chat_type=str(query_chat_type) if query_chat_type is not None else None,
-                    thread_id=str(query_thread_id) if query_thread_id is not None else None,
-                    user_name=query_user_name,
-                ):
+                caller_id = str(getattr(query.from_user, "id", "")) 
+                if not self._is_callback_user_authorized(caller_id):
                     await query.answer(text="⛔ You are not authorized to answer this prompt.")
                     return
 
@@ -1952,13 +1891,7 @@ class TelegramAdapter(BasePlatformAdapter):
             return
         answer = data.split(":", 1)[1]  # "y" or "n"
         caller_id = str(getattr(query.from_user, "id", ""))
-        if not self._is_callback_user_authorized(
-            caller_id,
-            chat_id=query_chat_id,
-            chat_type=str(query_chat_type) if query_chat_type is not None else None,
-            thread_id=str(query_thread_id) if query_thread_id is not None else None,
-            user_name=query_user_name,
-        ):
+        if not self._is_callback_user_authorized(caller_id):
             await query.answer(text="⛔ You are not authorized to answer update prompts.")
             return
         await query.answer(text=f"Sent '{answer}' to the update process.")

gateway/platforms/yuanbao.py

@@ -1896,12 +1896,10 @@ class OwnerCommandMiddleware(InboundMiddleware):
         if cmd not in cls.ALLOWLIST:
             return None, None, False
 
-        # Sender identity check: bot owner <-> push.from_account == push.bot_owner_id.
-        # The allowlisted commands (/approve, /deny, /stop, /reset, ...) are
-        # privileged — leaking them to non-owners lets any group member approve
-        # a dangerous tool call, kill the owner's task, or wipe session state.
-        owner_id = str((push or {}).get("bot_owner_id") or "").strip()
-        is_owner = bool(owner_id) and owner_id == from_account
+        # Sender identity check: bot owner <-> push.from_account == push.bot_owner_id
+        # owner_id = (push or {}).get("bot_owner_id") or ""
+        # is_owner = bool(owner_id) and owner_id == from_account
+        is_owner = True
         return cmd, cmd_line, is_owner
 
     async def handle(self, ctx: InboundContext, next_fn) -> None:

gateway/run.py

@@ -29,7 +29,7 @@ from collections import OrderedDict
 from contextvars import copy_context
 from pathlib import Path
 from datetime import datetime
-from typing import Dict, Optional, Any, List, Union
+from typing import Dict, Optional, Any, List
 
 # account_usage imports the OpenAI SDK chain (~230 ms). Only needed by
 # /usage; we still import it at module top in the gateway because test
@@ -232,16 +232,6 @@ def _ensure_ssl_certs() -> None:
             os.environ["SSL_CERT_FILE"] = candidate
             return
 
-def _home_target_env_var(platform_name: str) -> str:
-    """Return the configured home-target env var for a platform."""
-    from cron.scheduler import _HOME_TARGET_ENV_VARS
-
-    return _HOME_TARGET_ENV_VARS.get(
-        platform_name.lower(),
-        f"{platform_name.upper()}_HOME_CHANNEL",
-    )
-
-
 _ensure_ssl_certs()
 
 # Add parent directory to path
@@ -249,7 +239,7 @@ sys.path.insert(0, str(Path(__file__).parent.parent))
 
 # Resolve Hermes home directory (respects HERMES_HOME override)
 from hermes_constants import get_hermes_home
-from utils import atomic_json_write, atomic_yaml_write, base_url_host_matches, is_truthy_value
+from utils import atomic_yaml_write, base_url_host_matches, is_truthy_value
 _hermes_home = get_hermes_home()
 
 # Load environment variables from ~/.hermes/.env first.
@@ -391,8 +381,6 @@ if _config_path.exists():
         if _display_cfg and isinstance(_display_cfg, dict):
             if "busy_input_mode" in _display_cfg and "HERMES_GATEWAY_BUSY_INPUT_MODE" not in os.environ:
                 os.environ["HERMES_GATEWAY_BUSY_INPUT_MODE"] = str(_display_cfg["busy_input_mode"])
-            if "busy_ack_enabled" in _display_cfg and "HERMES_GATEWAY_BUSY_ACK_ENABLED" not in os.environ:
-                os.environ["HERMES_GATEWAY_BUSY_ACK_ENABLED"] = str(_display_cfg["busy_ack_enabled"])
         # Timezone: bridge config.yaml → HERMES_TIMEZONE env var.
         # HERMES_TIMEZONE from .env takes precedence (already in os.environ).
         _tz_cfg = _cfg.get("timezone", "")
@@ -464,7 +452,6 @@ from gateway.session import (
 from gateway.delivery import DeliveryRouter
 from gateway.platforms.base import (
     BasePlatformAdapter,
-    EphemeralReply,
     MessageEvent,
     MessageType,
     merge_pending_message_event,
@@ -895,7 +882,6 @@ class GatewayRunner:
         # /new and /reset.  /model and other mid-session operations
         # preserve the queue.
         self._queued_events: Dict[str, List[MessageEvent]] = {}
-        self._pending_native_image_paths_by_session: Dict[str, List[str]] = {}
         self._busy_ack_ts: Dict[str, float] = {}  # last busy-ack timestamp per session (debounce)
         self._session_run_generation: Dict[str, int] = {}
 
@@ -1753,10 +1739,7 @@ class GatewayRunner:
             if cfg_path.exists():
                 with open(cfg_path, encoding="utf-8") as _f:
                     cfg = _y.safe_load(_f) or {}
-                return is_truthy_value(
-                    cfg_get(cfg, "display", "show_reasoning"),
-                    default=False,
-                )
+                return bool(cfg_get(cfg, "display", "show_reasoning", default=False))
         except Exception:
             pass
         return False
@@ -1978,14 +1961,6 @@ class GatewayRunner:
             except Exception:
                 pass  # don't let interrupt failure block the ack
 
-        # Check if busy ack is disabled — skip sending but still process the input.
-        # Placed before debounce so we don't stamp a "last ack" timestamp that was
-        # never actually delivered.
-        busy_ack_enabled = os.environ.get("HERMES_GATEWAY_BUSY_ACK_ENABLED", "true").lower() == "true"
-        if not busy_ack_enabled:
-            logger.debug("Busy ack suppressed for session %s", session_key)
-            return True  # input still processed, just no ack sent
-
         # Debounce: only send an acknowledgment once every 30 seconds per session
         # to avoid spamming the user when they send multiple messages quickly
         _BUSY_ACK_COOLDOWN = 30
@@ -2270,7 +2245,7 @@ class GatewayRunner:
         # (they might become active again next restart)
 
         try:
-            atomic_json_write(path, new_counts, indent=None)
+            path.write_text(json.dumps(new_counts))
         except Exception:
             pass
 
@@ -2338,7 +2313,7 @@ class GatewayRunner:
             if session_key in counts:
                 del counts[session_key]
                 if counts:
-                    atomic_json_write(path, counts, indent=None)
+                    path.write_text(json.dumps(counts))
                 else:
                     path.unlink(missing_ok=True)
         except Exception:
@@ -3972,11 +3947,6 @@ class GatewayRunner:
             Platform.QQBOT: "QQ_ALLOW_ALL_USERS",
             Platform.YUANBAO: "YUANBAO_ALLOW_ALL_USERS",
         }
-        # Bots admitted by {PLATFORM}_ALLOW_BOTS bypass the human allowlist (#4466).
-        platform_allow_bots_map = {
-            Platform.DISCORD: "DISCORD_ALLOW_BOTS",
-            Platform.FEISHU: "FEISHU_ALLOW_BOTS",
-        }
 
         # Plugin platforms: check the registry for auth env var names
         if source.platform not in platform_env_map:
@@ -3996,9 +3966,14 @@ class GatewayRunner:
         if platform_allow_all_var and os.getenv(platform_allow_all_var, "").lower() in ("true", "1", "yes"):
             return True
 
-        if getattr(source, "is_bot", False):
-            allow_bots_var = platform_allow_bots_map.get(source.platform)
-            if allow_bots_var and os.getenv(allow_bots_var, "none").lower().strip() in ("mentions", "all"):
+        # Discord bot senders that passed the DISCORD_ALLOW_BOTS platform
+        # filter are already authorized at the platform level — skip the
+        # user allowlist. Without this, bot messages allowed by
+        # DISCORD_ALLOW_BOTS=mentions/all would be rejected here with
+        # "Unauthorized user" (fixes #4466).
+        if source.platform == Platform.DISCORD and getattr(source, "is_bot", False):
+            allow_bots = os.getenv("DISCORD_ALLOW_BOTS", "none").lower().strip()
+            if allow_bots in ("mentions", "all"):
                 return True
 
         # Discord role-based access (DISCORD_ALLOWED_ROLES): the adapter's
@@ -4483,7 +4458,7 @@ class GatewayRunner:
                     invalidation_reason="stop_command",
                 )
                 logger.info("STOP for session %s — agent interrupted, session lock released", _quick_key)
-                return EphemeralReply("⚡ Stopped. You can continue this session.")
+                return "⚡ Stopped. You can continue this session."
 
             # /reset and /new must bypass the running-agent guard so they
             # actually dispatch as commands instead of being queued as user
@@ -4605,17 +4580,6 @@ class GatewayRunner:
             if _cmd_def_inner and _cmd_def_inner.name == "kanban":
                 return await self._handle_kanban_command(event)
 
-            # /goal is safe mid-run for status/pause/clear (inspection and
-            # control-plane only — doesn't interrupt the running turn).
-            # Setting a new goal text mid-run is rejected with the same
-            # "wait or /stop" message as /model so we don't race a second
-            # continuation prompt against the current turn.
-            if _cmd_def_inner and _cmd_def_inner.name == "goal":
-                _goal_arg = (event.get_command_args() or "").strip().lower()
-                if not _goal_arg or _goal_arg in ("status", "pause", "resume", "clear", "stop", "done"):
-                    return await self._handle_goal_command(event)
-                return "Agent is running — use /goal status / pause / clear mid-run, or /stop before setting a new goal."
-
             # Session-level toggles that are safe to run mid-agent —
             # /yolo can unblock a pending approval prompt, /verbose cycles
             # the tool-progress display mode for the ongoing stream.
@@ -4699,7 +4663,7 @@ class GatewayRunner:
                     # Force-clean the sentinel so the session is unlocked.
                     self._release_running_agent_state(_quick_key)
                     logger.info("HARD STOP (pending) for session %s — sentinel cleared", _quick_key)
-                    return EphemeralReply("⚡ Force-stopped. The agent was still starting — session unlocked.")
+                    return "⚡ Force-stopped. The agent was still starting — session unlocked."
                 # Queue the message so it will be picked up after the
                 # agent starts.
                 adapter = self.adapters.get(source.platform)
@@ -4932,9 +4896,6 @@ class GatewayRunner:
             # at the end of this function so the rewritten text is sent
             # to the agent as a regular user turn.
 
-        if canonical == "goal":
-            return await self._handle_goal_command(event)
-
         if canonical == "voice":
             return await self._handle_voice_command(event)
 
@@ -5080,36 +5041,7 @@ class GatewayRunner:
         _run_generation = self._begin_session_run_generation(_quick_key)
 
         try:
-            _agent_result = await self._handle_message_with_agent(event, source, _quick_key, _run_generation)
-            # Goal continuation: after the agent returns a final response
-            # for this turn, check any standing /goal — the judge will
-            # either mark it done, pause it (budget), or enqueue a
-            # continuation prompt back through the adapter FIFO so the
-            # next turn makes more progress. Wrapped in try/except so a
-            # broken judge never breaks normal message handling.
-            try:
-                _final_text = ""
-                if isinstance(_agent_result, dict):
-                    _final_text = str(_agent_result.get("final_response") or "")
-                elif isinstance(_agent_result, str):
-                    _final_text = _agent_result
-                # Skip for empty responses (interrupted / errored) — the
-                # judge would almost always say "continue" and we'd loop
-                # on error. Let the user drive the next turn.
-                if _final_text.strip():
-                    try:
-                        session_entry = self.session_store.get_or_create_session(source)
-                    except Exception:
-                        session_entry = None
-                    if session_entry is not None:
-                        self._post_turn_goal_continuation(
-                            session_entry=session_entry,
-                            source=source,
-                            final_response=_final_text,
-                        )
-            except Exception as _goal_exc:
-                logger.debug("goal continuation hook failed: %s", _goal_exc)
-            return _agent_result
+            return await self._handle_message_with_agent(event, source, _quick_key, _run_generation)
         finally:
             # If _run_agent replaced the sentinel with a real agent and
             # then cleaned it up, this is a no-op.  If we exited early
@@ -5137,29 +5069,22 @@ class GatewayRunner:
         preprocessing pipeline so sender attribution, image enrichment, STT,
         document notes, reply context, and @ references all behave the same.
 
-        Side effect: buffers per-session native image paths when the active
-        model supports native vision AND the user has images attached. The
-        caller consumes and clears that session-scoped buffer at the
-        ``run_conversation`` site to build a multimodal user turn. When the
-        list is empty, the ``_enrich_message_with_vision`` text path has
-        already run and images are represented in-text.
+        Side effect: writes ``self._pending_native_image_paths`` to a list of
+        local image paths when the active model supports native vision AND
+        the user has images attached. The caller consumes and clears this
+        attribute at the ``run_conversation`` site to build a multimodal user
+        turn. When the list is empty, the ``_enrich_message_with_vision``
+        text path has already run and images are represented in-text.
         """
         history = history or []
         message_text = event.text or ""
-        _group_sessions_per_user = getattr(self.config, "group_sessions_per_user", True)
-        _thread_sessions_per_user = getattr(self.config, "thread_sessions_per_user", False)
-        # Use the same helper every other call site uses so the write key here
-        # matches the consume key at the run_conversation site — even if the
-        # session store overrides build_session_key's default behavior.
-        session_key = self._session_key_for_source(source)
-        # Reset only this session's per-call buffer; other sessions may be
-        # concurrently preparing multimodal turns on the same runner.
-        self._consume_pending_native_image_paths(session_key)
+        # Reset per-call buffer; set only when native routing is chosen.
+        self._pending_native_image_paths = []
 
         _is_shared_multi_user = is_shared_multi_user_session(
             source,
-            group_sessions_per_user=_group_sessions_per_user,
-            thread_sessions_per_user=_thread_sessions_per_user,
+            group_sessions_per_user=getattr(self.config, "group_sessions_per_user", True),
+            thread_sessions_per_user=getattr(self.config, "thread_sessions_per_user", False),
         )
         if _is_shared_multi_user and source.user_name:
             message_text = f"[{source.user_name}] {message_text}"
@@ -5180,11 +5105,7 @@ class GatewayRunner:
                 _img_mode = self._decide_image_input_mode()
                 if _img_mode == "native":
                     # Defer attachment to the run_conversation call site.
-                    pending_native = getattr(self, "_pending_native_image_paths_by_session", None)
-                    if pending_native is None:
-                        pending_native = {}
-                        self._pending_native_image_paths_by_session = pending_native
-                    pending_native[session_key] = list(image_paths)
+                    self._pending_native_image_paths = list(image_paths)
                     logger.info(
                         "Image routing: native (model supports vision). %d image(s) will be attached inline.",
                         len(image_paths),
@@ -5323,12 +5244,6 @@ class GatewayRunner:
 
         return message_text
 
-    def _consume_pending_native_image_paths(self, session_key: str) -> List[str]:
-        pending_native = getattr(self, "_pending_native_image_paths_by_session", None)
-        if not pending_native:
-            return []
-        return list(pending_native.pop(session_key, []) or [])
-
     async def _handle_message_with_agent(self, event, source, _quick_key: str, run_generation: int):
         """Inner handler that runs under the _running_agents sentinel guard."""
         _msg_start_time = time.time()
@@ -5357,12 +5272,7 @@ class GatewayRunner:
         _is_new_session = (
             session_entry.created_at == session_entry.updated_at
             or getattr(session_entry, "was_auto_reset", False)
-            or getattr(session_entry, "is_fresh_reset", False)
         )
-        # Consume the is_fresh_reset flag immediately so it doesn't leak
-        # onto subsequent messages in the same session (issue #6508).
-        if getattr(session_entry, "is_fresh_reset", False):
-            session_entry.is_fresh_reset = False
         if _is_new_session:
             await self.hooks.emit("session:start", {
                 "platform": source.platform.value if source.platform else "",
@@ -5811,7 +5721,7 @@ class GatewayRunner:
         # Skip for webhooks - they deliver directly to configured targets (github_comment, etc.)
         if not history and source.platform and source.platform != Platform.LOCAL and source.platform != Platform.WEBHOOK:
             platform_name = source.platform.value
-            env_key = _home_target_env_var(platform_name)
+            env_key = f"{platform_name.upper()}_HOME_CHANNEL"
             if not os.getenv(env_key):
                 adapter = self.adapters.get(source.platform)
                 if adapter:
@@ -6407,7 +6317,7 @@ class GatewayRunner:
 
         return "\n".join(lines)
 
-    async def _handle_reset_command(self, event: MessageEvent) -> Union[str, EphemeralReply]:
+    async def _handle_reset_command(self, event: MessageEvent) -> str:
         """Handle /new or /reset command."""
         source = event.source
         
@@ -6518,8 +6428,8 @@ class GatewayRunner:
             _tip_line = ""
 
         if session_info:
-            return EphemeralReply(f"{header}\n\n{session_info}{_tip_line}")
-        return EphemeralReply(f"{header}{_tip_line}")
+            return f"{header}\n\n{session_info}{_tip_line}"
+        return f"{header}{_tip_line}"
 
     async def _handle_profile_command(self, event: MessageEvent) -> str:
         """Handle /profile — show active profile name and home directory."""
@@ -6630,30 +6540,11 @@ class GatewayRunner:
         queue_depth = self._queue_depth(session_key, adapter=adapter)
 
         title = None
-        # Pull token totals from the SQLite session DB rather than the
-        # in-memory SessionStore.  The agent's per-turn token deltas are
-        # persisted into sessions_db (run_agent.py), not into SessionEntry,
-        # so session_entry.total_tokens is always 0.  SessionDB is the
-        # single source of truth; reading it here keeps /status accurate
-        # without duplicating token writes into two stores.
-        db_total_tokens = 0
         if self._session_db:
             try:
                 title = self._session_db.get_session_title(session_entry.session_id)
             except Exception:
                 title = None
-            try:
-                row = self._session_db.get_session(session_entry.session_id)
-                if row:
-                    db_total_tokens = (
-                        (row.get("input_tokens") or 0)
-                        + (row.get("output_tokens") or 0)
-                        + (row.get("cache_read_tokens") or 0)
-                        + (row.get("cache_write_tokens") or 0)
-                        + (row.get("reasoning_tokens") or 0)
-                    )
-            except Exception:
-                db_total_tokens = 0
 
         lines = [
             "📊 **Hermes Gateway Status**",
@@ -6665,7 +6556,7 @@ class GatewayRunner:
         lines.extend([
             f"**Created:** {session_entry.created_at.strftime('%Y-%m-%d %H:%M')}",
             f"**Last Activity:** {session_entry.updated_at.strftime('%Y-%m-%d %H:%M')}",
-            f"**Tokens:** {db_total_tokens:,}",
+            f"**Tokens:** {session_entry.total_tokens:,}",
             f"**Agent Running:** {'Yes ⚡' if is_running else 'No'}",
         ])
         if queue_depth:
@@ -6767,7 +6658,7 @@ class GatewayRunner:
 
         return "\n".join(lines)
 
-    async def _handle_stop_command(self, event: MessageEvent) -> Union[str, EphemeralReply]:
+    async def _handle_stop_command(self, event: MessageEvent) -> str:
         """Handle /stop command - interrupt a running agent.
 
         When an agent is truly hung (blocked thread that never checks
@@ -6792,7 +6683,7 @@ class GatewayRunner:
                 invalidation_reason="stop_command_pending",
             )
             logger.info("STOP (pending) for session %s — sentinel cleared", session_key)
-            return EphemeralReply("⚡ Stopped. The agent hadn't started yet — you can continue this session.")
+            return "⚡ Stopped. The agent hadn't started yet — you can continue this session."
         if agent:
             # Force-clean the session lock so a truly hung agent doesn't
             # keep it locked forever.
@@ -6802,11 +6693,11 @@ class GatewayRunner:
                 interrupt_reason=_INTERRUPT_REASON_STOP,
                 invalidation_reason="stop_command_handler",
             )
-            return EphemeralReply("⚡ Stopped. You can continue this session.")
+            return "⚡ Stopped. You can continue this session."
         else:
             return "No active task to stop."
 
-    async def _handle_restart_command(self, event: MessageEvent) -> Union[str, EphemeralReply]:
+    async def _handle_restart_command(self, event: MessageEvent) -> str:
         """Handle /restart command - drain active work, then restart the gateway."""
         # Defensive idempotency check: if the previous gateway process
         # recorded this same /restart (same platform + update_id) and the new
@@ -6832,7 +6723,7 @@ class GatewayRunner:
             count = self._running_agent_count()
             if count:
                 return f"⏳ Draining {count} active agent(s) before restart..."
-            return EphemeralReply("⏳ Gateway restart already in progress...")
+            return "⏳ Gateway restart already in progress..."
 
         # Save the requester's routing info so the new gateway process can
         # notify them once it comes back online.
@@ -6843,10 +6734,8 @@ class GatewayRunner:
             }
             if event.source.thread_id:
                 notify_data["thread_id"] = event.source.thread_id
-            atomic_json_write(
-                _hermes_home / ".restart_notify.json",
-                notify_data,
-                indent=None,
+            (_hermes_home / ".restart_notify.json").write_text(
+                json.dumps(notify_data)
             )
         except Exception as e:
             logger.debug("Failed to write restart notify file: %s", e)
@@ -6863,10 +6752,8 @@ class GatewayRunner:
             }
             if event.platform_update_id is not None:
                 dedup_data["update_id"] = event.platform_update_id
-            atomic_json_write(
-                _hermes_home / ".restart_last_processed.json",
-                dedup_data,
-                indent=None,
+            (_hermes_home / ".restart_last_processed.json").write_text(
+                json.dumps(dedup_data)
             )
         except Exception as e:
             logger.debug("Failed to write restart dedup marker: %s", e)
@@ -6884,7 +6771,7 @@ class GatewayRunner:
             self.request_restart(detached=True, via_service=False)
         if active_agents:
             return f"⏳ Draining {active_agents} active agent(s) before restart..."
-        return EphemeralReply("♻ Restarting gateway. If you aren't notified within 60 seconds, restart from the console with `hermes gateway restart`.")
+        return "♻ Restarting gateway. If you aren't notified within 60 seconds, restart from the console with `hermes gateway restart`."
 
     def _is_stale_restart_redelivery(self, event: MessageEvent) -> bool:
         """Return True if this /restart is a Telegram re-delivery we already handled.
@@ -7475,201 +7362,6 @@ class GatewayRunner:
         # Let the normal message handler process it
         return await self._handle_message(retry_event)
 
-    # ────────────────────────────────────────────────────────────────
-    # /goal — persistent cross-turn goals (Ralph-style loop)
-    # ────────────────────────────────────────────────────────────────
-    def _get_goal_manager_for_event(self, event: "MessageEvent"):
-        """Return a GoalManager bound to the session for this gateway event.
-
-        Returns ``(manager, session_entry)`` or ``(None, None)`` if the
-        goals module can't be loaded.
-        """
-        try:
-            from hermes_cli.goals import GoalManager
-        except Exception as exc:
-            logger.debug("goal manager unavailable: %s", exc)
-            return None, None
-        try:
-            session_entry = self.session_store.get_or_create_session(event.source)
-        except Exception as exc:
-            logger.debug("goal manager: session lookup failed: %s", exc)
-            return None, None
-        sid = getattr(session_entry, "session_id", None) or ""
-        if not sid:
-            return None, None
-        try:
-            goals_cfg = (
-                (self.config or {}).get("goals", {})
-                if isinstance(self.config, dict)
-                else getattr(self.config, "goals", {}) or {}
-            )
-            max_turns = int(goals_cfg.get("max_turns", 20) or 20)
-        except Exception:
-            max_turns = 20
-        return GoalManager(session_id=sid, default_max_turns=max_turns), session_entry
-
-    async def _handle_goal_command(self, event: "MessageEvent") -> str:
-        """Handle /goal for gateway platforms.
-
-        Subcommands: ``/goal`` / ``/goal status`` / ``/goal pause`` /
-        ``/goal resume`` / ``/goal clear``. Any other text becomes the
-        new goal.
-
-        Setting a new goal queues the goal text as the next turn so the
-        agent starts working on it immediately — the post-turn
-        continuation hook then takes over from there.
-        """
-        args = (event.get_command_args() or "").strip()
-        lower = args.lower()
-
-        mgr, session_entry = self._get_goal_manager_for_event(event)
-        if mgr is None:
-            return "Goals unavailable on this session."
-
-        if not args or lower == "status":
-            return mgr.status_line()
-
-        if lower == "pause":
-            state = mgr.pause(reason="user-paused")
-            if state is None:
-                return "No goal set."
-            return f"⏸ Goal paused: {state.goal}"
-
-        if lower == "resume":
-            state = mgr.resume()
-            if state is None:
-                return "No goal to resume."
-            return (
-                f"▶ Goal resumed: {state.goal}\n"
-                "Send any message to continue, or wait — I'll take the next step on the next turn."
-            )
-
-        if lower in ("clear", "stop", "done"):
-            had = mgr.has_goal()
-            mgr.clear()
-            return "✓ Goal cleared." if had else "No active goal."
-
-        # Otherwise — treat the remaining text as the new goal.
-        try:
-            state = mgr.set(args)
-        except ValueError as exc:
-            return f"Invalid goal: {exc}"
-
-        # Queue the goal text as an immediate first turn so the agent
-        # starts making progress. The post-turn hook takes over after.
-        adapter = self.adapters.get(event.source.platform) if event.source else None
-        _quick_key = self._session_key_for_source(event.source) if event.source else None
-        if adapter and _quick_key:
-            try:
-                kickoff_event = MessageEvent(
-                    text=state.goal,
-                    message_type=MessageType.TEXT,
-                    source=event.source,
-                    message_id=event.message_id,
-                    channel_prompt=event.channel_prompt,
-                )
-                self._enqueue_fifo(_quick_key, kickoff_event, adapter)
-            except Exception as exc:
-                logger.debug("goal kickoff enqueue failed: %s", exc)
-
-        return (
-            f"⊙ Goal set ({state.max_turns}-turn budget): {state.goal}\n"
-            "I'll keep working until the goal is done, you pause/clear it, or the budget is exhausted.\n"
-            "Controls: /goal status · /goal pause · /goal resume · /goal clear"
-        )
-
-    def _post_turn_goal_continuation(
-        self,
-        *,
-        session_entry: Any,
-        source: Any,
-        final_response: str,
-    ) -> None:
-        """Run the goal judge after a gateway turn and, if still active,
-        enqueue a continuation prompt for the same session.
-
-        Called from ``_handle_message_with_agent`` at turn boundary, AFTER
-        the response has been delivered. Safe when no goal is set.
-
-        We use the adapter's pending-message / FIFO machinery so any real
-        user message that arrives simultaneously is handled by the same
-        queue and takes priority naturally.
-        """
-        try:
-            from hermes_cli.goals import GoalManager
-        except Exception as exc:
-            logger.debug("goal continuation: goals module unavailable: %s", exc)
-            return
-
-        sid = getattr(session_entry, "session_id", None) or ""
-        if not sid:
-            return
-
-        try:
-            goals_cfg = (
-                (self.config or {}).get("goals", {})
-                if isinstance(self.config, dict)
-                else getattr(self.config, "goals", {}) or {}
-            )
-            max_turns = int(goals_cfg.get("max_turns", 20) or 20)
-        except Exception:
-            max_turns = 20
-
-        mgr = GoalManager(session_id=sid, default_max_turns=max_turns)
-        if not mgr.is_active():
-            return
-
-        decision = mgr.evaluate_after_turn(final_response or "", user_initiated=True)
-        msg = decision.get("message") or ""
-
-        # Send the status line back to the user so they see the judge's
-        # verdict. Fire-and-forget via the adapter.
-        if msg and source is not None:
-            try:
-                adapter = self.adapters.get(source.platform)
-                if adapter and hasattr(adapter, "send_message"):
-                    import asyncio as _asyncio
-                    coro = adapter.send_message(source, msg)
-                    if _asyncio.iscoroutine(coro):
-                        try:
-                            loop = _asyncio.get_event_loop()
-                            if loop.is_running():
-                                loop.create_task(coro)
-                            else:
-                                loop.run_until_complete(coro)
-                        except RuntimeError:
-                            # No event loop in this thread — schedule on the main one.
-                            try:
-                                _asyncio.run_coroutine_threadsafe(coro, self._loop)
-                            except Exception:
-                                pass
-            except Exception as exc:
-                logger.debug("goal continuation: status send failed: %s", exc)
-
-        if not decision.get("should_continue"):
-            return
-
-        prompt = decision.get("continuation_prompt") or ""
-        if not prompt or source is None:
-            return
-
-        # Enqueue via the adapter's FIFO so a user message already in
-        # flight preempts the continuation naturally.
-        try:
-            adapter = self.adapters.get(source.platform)
-            _quick_key = self._session_key_for_source(source)
-            if adapter and _quick_key:
-                cont_event = MessageEvent(
-                    text=prompt,
-                    message_type=MessageType.TEXT,
-                    source=source,
-                    message_id=None,
-                    channel_prompt=None,
-                )
-                self._enqueue_fifo(_quick_key, cont_event, adapter)
-        except Exception as exc:
-            logger.debug("goal continuation: enqueue failed: %s", exc)
-
     async def _handle_undo_command(self, event: MessageEvent) -> str:
         """Handle /undo command - remove the last user/assistant exchange."""
         source = event.source
@@ -7701,16 +7393,16 @@ class GatewayRunner:
         platform_name = source.platform.value if source.platform else "unknown"
         chat_id = source.chat_id
         chat_name = source.chat_name or chat_id
-
-        env_key = _home_target_env_var(platform_name)
-
+        
+        env_key = f"{platform_name.upper()}_HOME_CHANNEL"
+        
         # Save to .env so it persists across restarts
         try:
             from hermes_cli.config import save_env_value
             save_env_value(env_key, str(chat_id))
         except Exception as e:
             return f"Failed to save home channel: {e}"
-
+        
         return (
             f"✅ Home channel set to **{chat_name}** (ID: {chat_id}).\n"
             f"Cron jobs and cross-platform messages will be delivered here."
@@ -8282,8 +7974,6 @@ class GatewayRunner:
 
             from hermes_cli.tools_config import _get_platform_tools
             enabled_toolsets = sorted(_get_platform_tools(user_config, platform_key))
-            agent_cfg = user_config.get("agent") or {}
-            disabled_toolsets = agent_cfg.get("disabled_toolsets") or None
 
             pr = self._provider_routing
             max_iterations = int(os.getenv("HERMES_MAX_ITERATIONS", "90"))
@@ -8300,7 +7990,6 @@ class GatewayRunner:
                     quiet_mode=True,
                     verbose_logging=False,
                     enabled_toolsets=enabled_toolsets,
-                    disabled_toolsets=disabled_toolsets,
                     reasoning_config=reasoning_config,
                     service_tier=self._service_tier,
                     request_overrides=turn_route.get("request_overrides"),
@@ -8570,7 +8259,7 @@ class GatewayRunner:
             return f"⚡ ✓ Priority Processing: **{label}** (saved to config)\n_(takes effect on next message)_"
         return f"⚡ ✓ Priority Processing: **{label}** (this session only)"
 
-    async def _handle_yolo_command(self, event: MessageEvent) -> Union[str, EphemeralReply]:
+    async def _handle_yolo_command(self, event: MessageEvent) -> str:
         """Handle /yolo — toggle dangerous command approval bypass for this session only."""
         from tools.approval import (
             disable_session_yolo,
@@ -8582,10 +8271,10 @@ class GatewayRunner:
         current = is_session_yolo_enabled(session_key)
         if current:
             disable_session_yolo(session_key)
-            return EphemeralReply("⚠️ YOLO mode **OFF** for this session — dangerous commands will require approval.")
+            return "⚠️ YOLO mode **OFF** for this session — dangerous commands will require approval."
         else:
             enable_session_yolo(session_key)
-            return EphemeralReply("⚡ YOLO mode **ON** for this session — all commands auto-approved. Use with caution.")
+            return "⚡ YOLO mode **ON** for this session — all commands auto-approved. Use with caution."
 
     async def _handle_verbose_command(self, event: MessageEvent) -> str:
         """Handle /verbose command — cycle tool progress display mode.
@@ -8603,10 +8292,7 @@ class GatewayRunner:
         # --- check config gate ------------------------------------------------
         try:
             user_config = _load_gateway_config()
-            gate_enabled = is_truthy_value(
-                cfg_get(user_config, "display", "tool_progress_command"),
-                default=False,
-            )
+            gate_enabled = cfg_get(user_config, "display", "tool_progress_command", default=False)
         except Exception:
             gate_enabled = False
 
@@ -8761,7 +8447,7 @@ class GatewayRunner:
         try:
             from run_agent import AIAgent
             from agent.manual_compression_feedback import summarize_manual_compression
-            from agent.model_metadata import estimate_request_tokens_rough
+            from agent.model_metadata import estimate_messages_tokens_rough
 
             session_key = self._session_key_for_source(source)
             model, runtime_kwargs = self._resolve_session_agent_runtime(
@@ -8776,6 +8462,7 @@ class GatewayRunner:
                 for m in history
                 if m.get("role") in ("user", "assistant") and m.get("content")
             ]
+            approx_tokens = estimate_messages_tokens_rough(msgs)
 
             tmp_agent = AIAgent(
                 **runtime_kwargs,
@@ -8789,16 +8476,6 @@ class GatewayRunner:
             try:
                 tmp_agent._print_fn = lambda *a, **kw: None
 
-                # Estimate with system prompt + tool schemas included so the
-                # figure reflects real request pressure, not a transcript-only
-                # underestimate (#6217). Must be computed after tmp_agent is
-                # built so _cached_system_prompt/tools are populated.
-                _sys_prompt = getattr(tmp_agent, "_cached_system_prompt", "") or ""
-                _tools = getattr(tmp_agent, "tools", None) or None
-                approx_tokens = estimate_request_tokens_rough(
-                    msgs, system_prompt=_sys_prompt, tools=_tools
-                )
-
                 compressor = tmp_agent.context_compressor
                 if not compressor.has_content_to_compress(msgs):
                     return "Nothing to compress yet (the transcript is still all protected context)."
@@ -8823,9 +8500,7 @@ class GatewayRunner:
                 self.session_store.update_session(
                     session_entry.session_key, last_prompt_tokens=0
                 )
-                new_tokens = estimate_request_tokens_rough(
-                    compressed, system_prompt=_sys_prompt, tools=_tools
-                )
+                new_tokens = estimate_messages_tokens_rough(compressed)
                 summary = summarize_manual_compression(
                     msgs,
                     compressed,
@@ -9060,12 +8735,8 @@ class GatewayRunner:
                     tool_name=msg.get("tool_name") or msg.get("name"),
                     tool_calls=msg.get("tool_calls"),
                     tool_call_id=msg.get("tool_call_id"),
-                    finish_reason=msg.get("finish_reason"),
                     reasoning=msg.get("reasoning"),
                     reasoning_content=msg.get("reasoning_content"),
-                    reasoning_details=msg.get("reasoning_details"),
-                    codex_reasoning_items=msg.get("codex_reasoning_items"),
-                    codex_message_items=msg.get("codex_message_items"),
                 )
             except Exception:
                 pass  # Best-effort copy
@@ -9815,8 +9486,6 @@ class GatewayRunner:
             "session_key": session_key,
             "timestamp": datetime.now().isoformat(),
         }
-        if event.source.thread_id:
-            pending["thread_id"] = event.source.thread_id
         _tmp_pending = pending_path.with_suffix(".tmp")
         _tmp_pending.write_text(json.dumps(pending))
         _tmp_pending.replace(pending_path)
@@ -9902,7 +9571,6 @@ class GatewayRunner:
         adapter = None
         chat_id = None
         session_key = None
-        metadata = None
         for path in (claimed_path, pending_path):
             if path.exists():
                 try:
@@ -9910,8 +9578,6 @@ class GatewayRunner:
                     platform_str = pending.get("platform")
                     chat_id = pending.get("chat_id")
                     session_key = pending.get("session_key")
-                    thread_id = pending.get("thread_id")
-                    metadata = {"thread_id": thread_id} if thread_id else None
                     if platform_str and chat_id:
                         platform = Platform(platform_str)
                         adapter = self.adapters.get(platform)
@@ -9959,7 +9625,7 @@ class GatewayRunner:
             chunks = [clean[i:i + max_chunk] for i in range(0, len(clean), max_chunk)]
             for chunk in chunks:
                 try:
-                    await adapter.send(chat_id, f"```\n{chunk}\n```", metadata=metadata)
+                    await adapter.send(chat_id, f"```\n{chunk}\n```")
                 except Exception as e:
                     logger.debug("Update stream send failed: %s", e)
 
@@ -9982,13 +9648,9 @@ class GatewayRunner:
                     exit_code_raw = exit_code_path.read_text().strip() or "1"
                     exit_code = int(exit_code_raw)
                     if exit_code == 0:
-                        await adapter.send(chat_id, "✅ Hermes update finished.", metadata=metadata)
+                        await adapter.send(chat_id, "✅ Hermes update finished.")
                     else:
-                        await adapter.send(
-                            chat_id,
-                            "❌ Hermes update failed (exit code {}).".format(exit_code),
-                            metadata=metadata,
-                        )
+                        await adapter.send(chat_id, "❌ Hermes update failed (exit code {}).".format(exit_code))
                     logger.info("Update finished (exit=%s), notified %s", exit_code, session_key)
                 except Exception as e:
                     logger.warning("Update final notification failed: %s", e)
@@ -10038,7 +9700,6 @@ class GatewayRunner:
                                     prompt=prompt_text,
                                     default=default,
                                     session_key=session_key,
-                                    metadata=metadata,
                                 )
                                 sent_buttons = True
                             except Exception as btn_err:
@@ -10050,8 +9711,7 @@ class GatewayRunner:
                                 f"⚕ **Update needs your input:**\n\n"
                                 f"{prompt_text}{default_hint}\n\n"
                                 f"Reply `/approve` (yes) or `/deny` (no), "
-                                f"or type your answer directly.",
-                                metadata=metadata,
+                                f"or type your answer directly."
                             )
                         self._update_prompt_pending[session_key] = True
                         # Remove the prompt file so it isn't re-read on the
@@ -10071,11 +9731,7 @@ class GatewayRunner:
             exit_code_path.write_text("124")
             await _flush_buffer()
             try:
-                await adapter.send(
-                    chat_id,
-                    "❌ Hermes update timed out after 30 minutes.",
-                    metadata=metadata,
-                )
+                await adapter.send(chat_id, "❌ Hermes update timed out after 30 minutes.")
             except Exception:
                 pass
             for p in (pending_path, claimed_path, output_path,
@@ -10117,7 +9773,6 @@ class GatewayRunner:
             pending = json.loads(claimed_path.read_text())
             platform_str = pending.get("platform")
             chat_id = pending.get("chat_id")
-            thread_id = pending.get("thread_id")
 
             if not exit_code_path.exists():
                 logger.info("Update notification deferred: update still running")
@@ -10139,7 +9794,6 @@ class GatewayRunner:
             adapter = self.adapters.get(platform)
 
             if adapter and chat_id:
-                metadata = {"thread_id": thread_id} if thread_id else None
                 # Strip ANSI escape codes for clean display
                 output = re.sub(r'\x1b\[[0-9;]*m', '', output).strip()
                 if output:
@@ -10154,7 +9808,7 @@ class GatewayRunner:
                         msg = "✅ Hermes update finished successfully."
                     else:
                         msg = "❌ Hermes update failed. Check the gateway logs or run `hermes update` manually for details."
-                await adapter.send(chat_id, msg, metadata=metadata)
+                await adapter.send(chat_id, msg)
                 logger.info(
                     "Sent post-update notification to %s:%s (exit=%s)",
                     platform_str,
@@ -10694,7 +10348,6 @@ class GatewayRunner:
         ("compression", "threshold"),
         ("compression", "target_ratio"),
         ("compression", "protect_last_n"),
-        ("agent", "disabled_toolsets"),
     )
 
     @classmethod
@@ -11478,8 +11131,6 @@ class GatewayRunner:
 
         from hermes_cli.tools_config import _get_platform_tools
         enabled_toolsets = sorted(_get_platform_tools(user_config, platform_key))
-        agent_cfg_local = user_config.get("agent") or {}
-        disabled_toolsets = agent_cfg_local.get("disabled_toolsets") or None
 
         display_config = user_config.get("display", {})
         if not isinstance(display_config, dict):
@@ -11568,10 +11219,7 @@ class GatewayRunner:
                             tool_progress_hint_gateway,
                         )
                         _cfg = _load_gateway_config()
-                        gate_on = is_truthy_value(
-                            cfg_get(_cfg, "display", "tool_progress_command"),
-                            default=False,
-                        )
+                        gate_on = bool(cfg_get(_cfg, "display", "tool_progress_command", default=False))
                         if gate_on and not is_seen(_cfg, TOOL_PROGRESS_FLAG):
                             long_tool_hint_fired[0] = True
                             progress_queue.put(tool_progress_hint_gateway())
@@ -12111,7 +11759,6 @@ class GatewayRunner:
                     quiet_mode=True,
                     verbose_logging=False,
                     enabled_toolsets=enabled_toolsets,
-                    disabled_toolsets=disabled_toolsets,
                     ephemeral_system_prompt=combined_ephemeral or None,
                     prefill_messages=self._prefill_messages or None,
                     reasoning_config=reasoning_config,
@@ -12451,7 +12098,8 @@ class GatewayRunner:
                 # attachment, wrap the user turn as an OpenAI-style multimodal
                 # content list. Consume-and-clear so subsequent turns on the same
                 # runner instance don't re-attach stale images.
-                _native_imgs = self._consume_pending_native_image_paths(session_key)
+                _native_imgs = list(getattr(self, "_pending_native_image_paths", []) or [])
+                self._pending_native_image_paths = []
                 if _native_imgs:
                     try:
                         from agent.image_routing import build_native_content_parts

gateway/session.py

@@ -458,15 +458,6 @@ class SessionEntry:
     was_auto_reset: bool = False
     auto_reset_reason: Optional[str] = None  # "idle" or "daily"
     reset_had_activity: bool = False  # whether the expired session had any messages
-
-    # Set by reset_session() when the user explicitly sends /new or /reset.
-    # Consumed once by _handle_message_with_agent to trigger topic/channel
-    # skill re-injection on the first message of the new session.  We can't
-    # reuse was_auto_reset for this because that flag fires the "session
-    # expired due to inactivity" user-facing notice and a misleading
-    # context-note prepend — both wrong for an explicit manual reset.
-    # See issue #6508.
-    is_fresh_reset: bool = False
     
     # Set by the background expiry watcher after it finalizes an expired
     # session (invoking on_session_finalize hooks and evicting the cached
@@ -517,7 +508,6 @@ class SessionEntry:
                 if self.last_resume_marked_at
                 else None
             ),
-            "is_fresh_reset": self.is_fresh_reset,
         }
         if self.origin:
             result["origin"] = self.origin.to_dict()
@@ -566,7 +556,6 @@ class SessionEntry:
             resume_pending=data.get("resume_pending", False),
             resume_reason=data.get("resume_reason"),
             last_resume_marked_at=last_resume_marked_at,
-            is_fresh_reset=data.get("is_fresh_reset", False),
         )
 
 
@@ -1143,7 +1132,6 @@ class SessionStore:
                 display_name=old_entry.display_name,
                 platform=old_entry.platform,
                 chat_type=old_entry.chat_type,
-                is_fresh_reset=True,
             )
 
             self._entries[session_key] = new_entry

gateway/status.py

@@ -21,7 +21,6 @@ from datetime import datetime, timezone
 from pathlib import Path
 from hermes_constants import get_hermes_home
 from typing import Any, Optional
-from utils import atomic_json_write
 
 if sys.platform == "win32":
     import msvcrt
@@ -35,10 +34,6 @@ _IS_WINDOWS = sys.platform == "win32"
 _UNSET = object()
 _GATEWAY_LOCK_FILENAME = "gateway.lock"
 _gateway_lock_handle = None
-# Windows byte-range locks are mandatory for other readers. Lock a byte well
-# past the JSON payload so runtime status / PID readers can still read the file
-# while another process holds the mutual-exclusion lock.
-_WINDOWS_LOCK_OFFSET = 1024 * 1024
 
 
 def _get_pid_path() -> Path:
@@ -210,7 +205,8 @@ def _read_json_file(path: Path) -> Optional[dict[str, Any]]:
 
 
 def _write_json_file(path: Path, payload: dict[str, Any]) -> None:
-    atomic_json_write(path, payload, indent=None, separators=(",", ":"))
+    path.parent.mkdir(parents=True, exist_ok=True)
+    path.write_text(json.dumps(payload))
 
 
 def _read_pid_record(pid_path: Optional[Path] = None) -> Optional[dict]:
@@ -290,7 +286,7 @@ def _try_acquire_file_lock(handle) -> bool:
             if handle.tell() == 0:
                 handle.write("\n")
                 handle.flush()
-            handle.seek(_WINDOWS_LOCK_OFFSET)
+            handle.seek(0)
             msvcrt.locking(handle.fileno(), msvcrt.LK_NBLCK, 1)
         else:
             fcntl.flock(handle.fileno(), fcntl.LOCK_EX | fcntl.LOCK_NB)
@@ -302,7 +298,7 @@ def _try_acquire_file_lock(handle) -> bool:
 def _release_file_lock(handle) -> None:
     try:
         if _IS_WINDOWS:
-            handle.seek(_WINDOWS_LOCK_OFFSET)
+            handle.seek(0)
             msvcrt.locking(handle.fileno(), msvcrt.LK_UNLCK, 1)
         else:
             fcntl.flock(handle.fileno(), fcntl.LOCK_UN)

hermes_cli/auth.py

@@ -43,7 +43,7 @@ import yaml
 
 from hermes_cli.config import get_hermes_home, get_config_path, read_raw_config
 from hermes_constants import OPENROUTER_BASE_URL
-from utils import atomic_replace, atomic_yaml_write, is_truthy_value
+from utils import atomic_replace
 
 logger = logging.getLogger(__name__)
 
@@ -2480,8 +2480,8 @@ def _resolve_verify(
     tls_state = tls_state if isinstance(tls_state, dict) else {}
 
     effective_insecure = (
-        is_truthy_value(insecure, default=False) if insecure is not None
-        else is_truthy_value(tls_state.get("insecure", False), default=False)
+        bool(insecure) if insecure is not None
+        else bool(tls_state.get("insecure", False))
     )
     effective_ca = (
         ca_bundle
@@ -3653,7 +3653,7 @@ def _update_config_for_provider(
 
     config["model"] = model_cfg
 
-    atomic_yaml_write(config_path, config, sort_keys=False)
+    config_path.write_text(yaml.safe_dump(config, sort_keys=False))
     return config_path
 
 
@@ -3712,7 +3712,7 @@ def _reset_config_provider() -> Path:
         model["provider"] = "auto"
         if "base_url" in model:
             model["base_url"] = OPENROUTER_BASE_URL
-    atomic_yaml_write(config_path, config, sort_keys=False)
+    config_path.write_text(yaml.safe_dump(config, sort_keys=False))
     return config_path
 
 

hermes_cli/commands.py

@@ -19,8 +19,6 @@ from collections.abc import Callable, Mapping
 from dataclasses import dataclass
 from typing import Any
 
-from utils import is_truthy_value
-
 # prompt_toolkit is an optional CLI dependency — only needed for
 # SlashCommandCompleter and SlashCommandAutoSuggest.  Gateway and test
 # environments that lack it must still be able to import this module
@@ -95,8 +93,6 @@ COMMAND_REGISTRY: list[CommandDef] = [
                aliases=("q",), args_hint="<prompt>"),
     CommandDef("steer", "Inject a message after the next tool call without interrupting", "Session",
                args_hint="<prompt>"),
-    CommandDef("goal", "Set a standing goal Hermes works on across turns until achieved", "Session",
-               args_hint="[text | pause | resume | clear | status]"),
     CommandDef("status", "Show session info", "Session"),
     CommandDef("profile", "Show active profile name and home directory", "Info"),
     CommandDef("sethome", "Set this chat as the home channel", "Session",
@@ -375,7 +371,7 @@ def _resolve_config_gates() -> set[str]:
             else:
                 val = None
                 break
-        if is_truthy_value(val, default=False):
+        if val:
             result.add(cmd.name)
     return result
 

hermes_cli/config.py

@@ -457,7 +457,6 @@ DEFAULT_CONFIG = {
         # remains available as a tool regardless of this setting — the routing
         # only controls how inbound user images are presented.
         "image_input_mode": "auto",
-        "disabled_toolsets": [],
     },
     
     "terminal": {
@@ -607,24 +606,6 @@ DEFAULT_CONFIG = {
         "max_line_length": 2000,
     },
 
-    # Tool loop guardrails nudge models when they repeat failed or
-    # non-progressing tool calls. Soft warnings are always-on by default;
-    # hard stops are opt-in so interactive CLI/TUI sessions keep flowing.
-    "tool_loop_guardrails": {
-        "warnings_enabled": True,
-        "hard_stop_enabled": False,
-        "warn_after": {
-            "exact_failure": 2,
-            "same_tool_failure": 3,
-            "idempotent_no_progress": 2,
-        },
-        "hard_stop_after": {
-            "exact_failure": 5,
-            "same_tool_failure": 8,
-            "idempotent_no_progress": 5,
-        },
-    },
-
     "compression": {
         "enabled": True,
         "threshold": 0.50,            # compress when context usage exceeds this ratio
@@ -775,14 +756,6 @@ DEFAULT_CONFIG = {
         "tool_progress_command": False,  # Enable /verbose command in messaging gateway
         "tool_progress_overrides": {},  # DEPRECATED — use display.platforms instead
         "tool_preview_length": 0,  # Max chars for tool call previews (0 = no limit, show full paths/commands)
-        # Auto-delete system-notice replies (e.g. "✨ New session started!",
-        # "♻ Restarting gateway…", "⚡ Stopped…") after N seconds on platforms
-        # that support message deletion (currently Telegram; other platforms
-        # ignore and leave the message in place).  Only affects slash-command
-        # replies wrapped with gateway.platforms.base.EphemeralReply — agent
-        # responses and content messages are never touched.  Default 0
-        # (disabled) preserves prior behavior.
-        "ephemeral_system_ttl": 0,
         "platforms": {},  # Per-platform display overrides: {"telegram": {"tool_progress": "all"}, "slack": {"tool_progress": "off"}}
         # Gateway runtime-metadata footer appended to the FINAL message of a turn
         # (disabled by default to keep replies minimal). When enabled, renders
@@ -952,23 +925,7 @@ DEFAULT_CONFIG = {
     # injected at the start of every API call for few-shot priming.
     # Never saved to sessions, logs, or trajectories.
     "prefill_messages_file": "",
-
-    # Goals — persistent cross-turn goals (Ralph-style loop).
-    # After every turn, a lightweight judge call asks the auxiliary model
-    # whether the active /goal is satisfied by the assistant's last
-    # response. If not, Hermes feeds a continuation prompt back into the
-    # same session and keeps working until the goal is done, the turn
-    # budget is exhausted, or the user pauses/clears it. Judge failures
-    # fail OPEN (continue) so a flaky judge never wedges progress — the
-    # turn budget is the real backstop.
-    "goals": {
-        # Max continuation turns before Hermes auto-pauses the goal and
-        # asks the user to /goal resume. Protects against judge false
-        # negatives (goal actually done but judge says continue) and
-        # unbounded model spend on fuzzy / unachievable goals.
-        "max_turns": 20,
-    },
-
+    
     # Skills — external skill directories for sharing skills across tools/agents.
     # Each path is expanded (~, ${VAR}) and resolved.  Read-only — skill creation
     # always goes to ~/.hermes/skills/.
@@ -2461,17 +2418,7 @@ def get_missing_skill_config_vars() -> List[Dict[str, Any]]:
     except Exception:
         return []
 
-    try:
-        all_vars = discover_all_skill_config_vars()
-    except Exception as e:
-        # A malformed SKILL.md, unreadable external skill dir, or similar
-        # should never break `hermes update`.  Skill-config prompting is a
-        # post-migration nicety, not a blocker.
-        import logging
-        logging.getLogger(__name__).debug(
-            "discover_all_skill_config_vars failed: %s", e
-        )
-        return []
+    all_vars = discover_all_skill_config_vars()
     if not all_vars:
         return []
 

hermes_cli/gateway.py

@@ -10,7 +10,6 @@ import shutil
 import signal
 import subprocess
 import sys
-import textwrap
 from dataclasses import dataclass
 from pathlib import Path
 
@@ -60,13 +59,6 @@ class GatewayRuntimeSnapshot:
     def has_process_service_mismatch(self) -> bool:
         return self.service_installed and self.running and not self.service_running
 
-
-@dataclass(frozen=True)
-class ProfileGatewayProcess:
-    profile: str
-    path: Path
-    pid: int
-
 def _get_service_pids() -> set:
     """Return PIDs currently managed by systemd or launchd gateway services.
 
@@ -379,83 +371,6 @@ def find_gateway_pids(exclude_pids: set | None = None, all_profiles: bool = Fals
     return pids
 
 
-def find_profile_gateway_processes(
-    exclude_pids: set | None = None,
-) -> list[ProfileGatewayProcess]:
-    """Return running gateway PIDs mapped to Hermes profiles via PID files."""
-    _exclude = set(exclude_pids or set())
-    processes: list[ProfileGatewayProcess] = []
-    try:
-        from gateway.status import get_running_pid
-        from hermes_cli.profiles import list_profiles
-    except Exception:
-        return processes
-
-    seen: set[int] = set()
-    for profile in list_profiles():
-        try:
-            pid = get_running_pid(profile.path / "gateway.pid", cleanup_stale=False)
-        except Exception:
-            continue
-        if pid is None or pid <= 0 or pid in _exclude or pid in seen:
-            continue
-        seen.add(pid)
-        processes.append(ProfileGatewayProcess(profile=profile.name, path=profile.path, pid=pid))
-    return processes
-
-
-def _gateway_run_args_for_profile(profile: str) -> list[str]:
-    args = [get_python_path(), "-m", "hermes_cli.main"]
-    if profile != "default":
-        args.extend(["--profile", profile])
-    args.extend(["gateway", "run", "--replace"])
-    return args
-
-
-def launch_detached_profile_gateway_restart(profile: str, old_pid: int) -> bool:
-    """Relaunch a manually-run profile gateway after its current PID exits."""
-    if old_pid <= 0:
-        return False
-
-    watcher = textwrap.dedent(
-        """
-        import os
-        import subprocess
-        import sys
-        import time
-
-        pid = int(sys.argv[1])
-        cmd = sys.argv[2:]
-        deadline = time.monotonic() + 120
-        while time.monotonic() < deadline:
-            try:
-                os.kill(pid, 0)
-            except ProcessLookupError:
-                break
-            except PermissionError:
-                pass
-            time.sleep(0.2)
-        subprocess.Popen(
-            cmd,
-            stdout=subprocess.DEVNULL,
-            stderr=subprocess.DEVNULL,
-            start_new_session=True,
-        )
-        """
-    ).strip()
-
-    try:
-        subprocess.Popen(
-            [sys.executable, "-c", watcher, str(old_pid), *_gateway_run_args_for_profile(profile)],
-            stdout=subprocess.DEVNULL,
-            stderr=subprocess.DEVNULL,
-            start_new_session=True,
-        )
-    except OSError:
-        return False
-    return True
-
-
 def _probe_systemd_service_running(system: bool = False) -> tuple[bool, bool]:
     selected_system = _select_systemd_scope(system)
     unit_exists = get_systemd_unit_path(system=selected_system).exists()
@@ -4462,4 +4377,4 @@ def _gateway_command_inner(args):
         if not supports_systemd_services() and not is_macos():
             print("Legacy unit migration only applies to systemd-based Linux hosts.")
             return
-        remove_legacy_hermes_units(interactive=not yes, dry_run=dry_run)
+        remove_legacy_hermes_units(interactive=not yes, dry_run=dry_run)

hermes_cli/goals.py

@@ -1,535 +0,0 @@
-"""Persistent session goals — the Ralph loop for Hermes.
-
-A goal is a free-form user objective that stays active across turns. After
-each turn completes, a small judge call asks an auxiliary model "is this
-goal satisfied by the assistant's last response?". If not, Hermes feeds a
-continuation prompt back into the same session and keeps working until the
-goal is done, turn budget is exhausted, the user pauses/clears it, or the
-user sends a new message (which takes priority and pauses the goal loop).
-
-State is persisted in SessionDB's ``state_meta`` table keyed by
-``goal:<session_id>`` so ``/resume`` picks it up.
-
-Design notes / invariants:
-
-- The continuation prompt is just a normal user message appended to the
-  session via ``run_conversation``. No system-prompt mutation, no toolset
-  swap — prompt caching stays intact.
-- Judge failures are fail-OPEN: ``continue``. A broken judge must not wedge
-  progress; the turn budget is the backstop.
-- When a real user message arrives mid-loop it preempts the continuation
-  prompt and also pauses the goal loop for that turn (we still re-judge
-  after, so if the user's message happens to complete the goal the judge
-  will say ``done``).
-- This module has zero hard dependency on ``cli.HermesCLI`` or the gateway
-  runner — both wire the same ``GoalManager`` in.
-
-Nothing in this module touches the agent's system prompt or toolset.
-"""
-
-from __future__ import annotations
-
-import json
-import logging
-import re
-import time
-from dataclasses import dataclass, asdict
-from typing import Any, Dict, Optional, Tuple
-
-logger = logging.getLogger(__name__)
-
-
-# ──────────────────────────────────────────────────────────────────────
-# Constants & defaults
-# ──────────────────────────────────────────────────────────────────────
-
-DEFAULT_MAX_TURNS = 20
-DEFAULT_JUDGE_TIMEOUT = 30.0
-# Cap how much of the last response + recent messages we send to the judge.
-_JUDGE_RESPONSE_SNIPPET_CHARS = 4000
-
-
-CONTINUATION_PROMPT_TEMPLATE = (
-    "[Continuing toward your standing goal]\n"
-    "Goal: {goal}\n\n"
-    "Continue working toward this goal. Take the next concrete step. "
-    "If you believe the goal is complete, state so explicitly and stop. "
-    "If you are blocked and need input from the user, say so clearly and stop."
-)
-
-
-JUDGE_SYSTEM_PROMPT = (
-    "You are a strict judge evaluating whether an autonomous agent has "
-    "achieved a user's stated goal. You receive the goal text and the "
-    "agent's most recent response. Your only job is to decide whether "
-    "the goal is fully satisfied based on that response.\n\n"
-    "A goal is DONE only when:\n"
-    "- The response explicitly confirms the goal was completed, OR\n"
-    "- The response clearly shows the final deliverable was produced, OR\n"
-    "- The response explains the goal is unachievable / blocked / needs "
-    "user input (treat this as DONE with reason describing the block).\n\n"
-    "Otherwise the goal is NOT done — CONTINUE.\n\n"
-    "Reply ONLY with a single JSON object on one line:\n"
-    '{\"done\": <true|false>, \"reason\": \"<one-sentence rationale>\"}'
-)
-
-
-JUDGE_USER_PROMPT_TEMPLATE = (
-    "Goal:\n{goal}\n\n"
-    "Agent's most recent response:\n{response}\n\n"
-    "Is the goal satisfied?"
-)
-
-
-# ──────────────────────────────────────────────────────────────────────
-# Dataclass
-# ──────────────────────────────────────────────────────────────────────
-
-
-@dataclass
-class GoalState:
-    """Serializable goal state stored per session."""
-
-    goal: str
-    status: str = "active"          # active | paused | done | cleared
-    turns_used: int = 0
-    max_turns: int = DEFAULT_MAX_TURNS
-    created_at: float = 0.0
-    last_turn_at: float = 0.0
-    last_verdict: Optional[str] = None        # "done" | "continue" | "skipped"
-    last_reason: Optional[str] = None
-    paused_reason: Optional[str] = None       # why we auto-paused (budget, etc.)
-
-    def to_json(self) -> str:
-        return json.dumps(asdict(self), ensure_ascii=False)
-
-    @classmethod
-    def from_json(cls, raw: str) -> "GoalState":
-        data = json.loads(raw)
-        return cls(
-            goal=data.get("goal", ""),
-            status=data.get("status", "active"),
-            turns_used=int(data.get("turns_used", 0) or 0),
-            max_turns=int(data.get("max_turns", DEFAULT_MAX_TURNS) or DEFAULT_MAX_TURNS),
-            created_at=float(data.get("created_at", 0.0) or 0.0),
-            last_turn_at=float(data.get("last_turn_at", 0.0) or 0.0),
-            last_verdict=data.get("last_verdict"),
-            last_reason=data.get("last_reason"),
-            paused_reason=data.get("paused_reason"),
-        )
-
-
-# ──────────────────────────────────────────────────────────────────────
-# Persistence (SessionDB state_meta)
-# ──────────────────────────────────────────────────────────────────────
-
-
-def _meta_key(session_id: str) -> str:
-    return f"goal:{session_id}"
-
-
-_DB_CACHE: Dict[str, Any] = {}
-
-
-def _get_session_db() -> Optional[Any]:
-    """Return a SessionDB instance for the current HERMES_HOME.
-
-    SessionDB has no built-in singleton, but opening a new connection per
-    /goal call would thrash the file. We cache one instance per
-    ``hermes_home`` path so profile switches still pick up the right DB.
-    Defensive against import/instantiation failures so tests and
-    non-standard launchers can still use the GoalManager.
-    """
-    try:
-        from hermes_constants import get_hermes_home
-        from hermes_state import SessionDB
-
-        home = str(get_hermes_home())
-    except Exception as exc:  # pragma: no cover
-        logger.debug("GoalManager: SessionDB bootstrap failed (%s)", exc)
-        return None
-
-    cached = _DB_CACHE.get(home)
-    if cached is not None:
-        return cached
-    try:
-        db = SessionDB()
-    except Exception as exc:  # pragma: no cover
-        logger.debug("GoalManager: SessionDB() raised (%s)", exc)
-        return None
-    _DB_CACHE[home] = db
-    return db
-
-
-def load_goal(session_id: str) -> Optional[GoalState]:
-    """Load the goal for a session, or None if none exists."""
-    if not session_id:
-        return None
-    db = _get_session_db()
-    if db is None:
-        return None
-    try:
-        raw = db.get_meta(_meta_key(session_id))
-    except Exception as exc:
-        logger.debug("GoalManager: get_meta failed: %s", exc)
-        return None
-    if not raw:
-        return None
-    try:
-        return GoalState.from_json(raw)
-    except Exception as exc:
-        logger.warning("GoalManager: could not parse stored goal for %s: %s", session_id, exc)
-        return None
-
-
-def save_goal(session_id: str, state: GoalState) -> None:
-    """Persist a goal to SessionDB. No-op if DB unavailable."""
-    if not session_id:
-        return
-    db = _get_session_db()
-    if db is None:
-        return
-    try:
-        db.set_meta(_meta_key(session_id), state.to_json())
-    except Exception as exc:
-        logger.debug("GoalManager: set_meta failed: %s", exc)
-
-
-def clear_goal(session_id: str) -> None:
-    """Mark a goal cleared in the DB (preserved for audit, status=cleared)."""
-    state = load_goal(session_id)
-    if state is None:
-        return
-    state.status = "cleared"
-    save_goal(session_id, state)
-
-
-# ──────────────────────────────────────────────────────────────────────
-# Judge
-# ──────────────────────────────────────────────────────────────────────
-
-
-def _truncate(text: str, limit: int) -> str:
-    if not text:
-        return ""
-    if len(text) <= limit:
-        return text
-    return text[:limit] + "… [truncated]"
-
-
-_JSON_OBJECT_RE = re.compile(r"\{.*?\}", re.DOTALL)
-
-
-def _parse_judge_response(raw: str) -> Tuple[bool, str]:
-    """Parse the judge's reply. Fail-open to ``(False, "<reason>")``.
-
-    Returns ``(done, reason)``.
-    """
-    if not raw:
-        return False, "judge returned empty response"
-
-    text = raw.strip()
-
-    # Strip markdown code fences the model may wrap JSON in.
-    if text.startswith("```"):
-        text = text.strip("`")
-        # Peel off leading json/JSON/etc tag
-        nl = text.find("\n")
-        if nl != -1:
-            text = text[nl + 1:]
-
-    # First try: parse the whole blob.
-    data: Optional[Dict[str, Any]] = None
-    try:
-        data = json.loads(text)
-    except Exception:
-        # Second try: pull the first JSON object out.
-        match = _JSON_OBJECT_RE.search(text)
-        if match:
-            try:
-                data = json.loads(match.group(0))
-            except Exception:
-                data = None
-
-    if not isinstance(data, dict):
-        return False, f"judge reply was not JSON: {_truncate(raw, 200)!r}"
-
-    done_val = data.get("done")
-    if isinstance(done_val, str):
-        done = done_val.strip().lower() in ("true", "yes", "1", "done")
-    else:
-        done = bool(done_val)
-    reason = str(data.get("reason") or "").strip()
-    if not reason:
-        reason = "no reason provided"
-    return done, reason
-
-
-def judge_goal(
-    goal: str,
-    last_response: str,
-    *,
-    timeout: float = DEFAULT_JUDGE_TIMEOUT,
-) -> Tuple[str, str]:
-    """Ask the auxiliary model whether the goal is satisfied.
-
-    Returns ``(verdict, reason)`` where verdict is ``"done"``, ``"continue"``,
-    or ``"skipped"`` (when the judge couldn't be reached).
-
-    This is deliberately fail-open: any error returns ``("continue", "...")``
-    so a broken judge doesn't wedge progress — the turn budget is the
-    backstop.
-    """
-    if not goal.strip():
-        return "skipped", "empty goal"
-    if not last_response.strip():
-        # No substantive reply this turn — almost certainly not done yet.
-        return "continue", "empty response (nothing to evaluate)"
-
-    try:
-        from agent.auxiliary_client import get_text_auxiliary_client
-    except Exception as exc:
-        logger.debug("goal judge: auxiliary client import failed: %s", exc)
-        return "continue", "auxiliary client unavailable"
-
-    try:
-        client, model = get_text_auxiliary_client("goal_judge")
-    except Exception as exc:
-        logger.debug("goal judge: get_text_auxiliary_client failed: %s", exc)
-        return "continue", "auxiliary client unavailable"
-
-    if client is None or not model:
-        return "continue", "no auxiliary client configured"
-
-    prompt = JUDGE_USER_PROMPT_TEMPLATE.format(
-        goal=_truncate(goal, 2000),
-        response=_truncate(last_response, _JUDGE_RESPONSE_SNIPPET_CHARS),
-    )
-
-    try:
-        resp = client.chat.completions.create(
-            model=model,
-            messages=[
-                {"role": "system", "content": JUDGE_SYSTEM_PROMPT},
-                {"role": "user", "content": prompt},
-            ],
-            temperature=0,
-            max_tokens=200,
-            timeout=timeout,
-        )
-    except Exception as exc:
-        logger.info("goal judge: API call failed (%s) — falling through to continue", exc)
-        return "continue", f"judge error: {type(exc).__name__}"
-
-    try:
-        raw = resp.choices[0].message.content or ""
-    except Exception:
-        raw = ""
-
-    done, reason = _parse_judge_response(raw)
-    verdict = "done" if done else "continue"
-    logger.info("goal judge: verdict=%s reason=%s", verdict, _truncate(reason, 120))
-    return verdict, reason
-
-
-# ──────────────────────────────────────────────────────────────────────
-# GoalManager — the orchestration surface CLI + gateway talk to
-# ──────────────────────────────────────────────────────────────────────
-
-
-class GoalManager:
-    """Per-session goal state + continuation decisions.
-
-    The CLI and gateway each hold one ``GoalManager`` per live session.
-
-    Methods:
-
-    - ``set(goal)`` — start a new standing goal.
-    - ``clear()`` — remove the active goal.
-    - ``pause()`` / ``resume()`` — explicit user controls.
-    - ``status()`` — printable one-liner.
-    - ``evaluate_after_turn(last_response)`` — call the judge, update state,
-      and return a decision dict the caller uses to drive the next turn.
-    - ``next_continuation_prompt()`` — the canonical user-role message to
-      feed back into ``run_conversation``.
-    """
-
-    def __init__(self, session_id: str, *, default_max_turns: int = DEFAULT_MAX_TURNS):
-        self.session_id = session_id
-        self.default_max_turns = int(default_max_turns or DEFAULT_MAX_TURNS)
-        self._state: Optional[GoalState] = load_goal(session_id)
-
-    # --- introspection ------------------------------------------------
-
-    @property
-    def state(self) -> Optional[GoalState]:
-        return self._state
-
-    def is_active(self) -> bool:
-        return self._state is not None and self._state.status == "active"
-
-    def has_goal(self) -> bool:
-        return self._state is not None and self._state.status in ("active", "paused")
-
-    def status_line(self) -> str:
-        s = self._state
-        if s is None or s.status in ("cleared",):
-            return "No active goal. Set one with /goal <text>."
-        turns = f"{s.turns_used}/{s.max_turns} turns"
-        if s.status == "active":
-            return f"⊙ Goal (active, {turns}): {s.goal}"
-        if s.status == "paused":
-            extra = f" — {s.paused_reason}" if s.paused_reason else ""
-            return f"⏸ Goal (paused, {turns}{extra}): {s.goal}"
-        if s.status == "done":
-            return f"✓ Goal done ({turns}): {s.goal}"
-        return f"Goal ({s.status}, {turns}): {s.goal}"
-
-    # --- mutation -----------------------------------------------------
-
-    def set(self, goal: str, *, max_turns: Optional[int] = None) -> GoalState:
-        goal = (goal or "").strip()
-        if not goal:
-            raise ValueError("goal text is empty")
-        state = GoalState(
-            goal=goal,
-            status="active",
-            turns_used=0,
-            max_turns=int(max_turns) if max_turns else self.default_max_turns,
-            created_at=time.time(),
-            last_turn_at=0.0,
-        )
-        self._state = state
-        save_goal(self.session_id, state)
-        return state
-
-    def pause(self, reason: str = "user-paused") -> Optional[GoalState]:
-        if not self._state:
-            return None
-        self._state.status = "paused"
-        self._state.paused_reason = reason
-        save_goal(self.session_id, self._state)
-        return self._state
-
-    def resume(self, *, reset_budget: bool = True) -> Optional[GoalState]:
-        if not self._state:
-            return None
-        self._state.status = "active"
-        self._state.paused_reason = None
-        if reset_budget:
-            self._state.turns_used = 0
-        save_goal(self.session_id, self._state)
-        return self._state
-
-    def clear(self) -> None:
-        if self._state is None:
-            return
-        self._state.status = "cleared"
-        save_goal(self.session_id, self._state)
-        self._state = None
-
-    def mark_done(self, reason: str) -> None:
-        if not self._state:
-            return
-        self._state.status = "done"
-        self._state.last_verdict = "done"
-        self._state.last_reason = reason
-        save_goal(self.session_id, self._state)
-
-    # --- the main entry point called after every turn -----------------
-
-    def evaluate_after_turn(
-        self,
-        last_response: str,
-        *,
-        user_initiated: bool = True,
-    ) -> Dict[str, Any]:
-        """Run the judge and update state. Return a decision dict.
-
-        ``user_initiated`` distinguishes a real user prompt (True) from a
-        continuation prompt we fed ourselves (False). Both increment
-        ``turns_used`` because both consume model budget.
-
-        Decision keys:
-          - ``status``: current goal status after update
-          - ``should_continue``: bool — caller should fire another turn
-          - ``continuation_prompt``: str or None
-          - ``verdict``: "done" | "continue" | "skipped" | "inactive"
-          - ``reason``: str
-          - ``message``: user-visible one-liner to print/send
-        """
-        state = self._state
-        if state is None or state.status != "active":
-            return {
-                "status": state.status if state else None,
-                "should_continue": False,
-                "continuation_prompt": None,
-                "verdict": "inactive",
-                "reason": "no active goal",
-                "message": "",
-            }
-
-        # Count the turn that just finished.
-        state.turns_used += 1
-        state.last_turn_at = time.time()
-
-        verdict, reason = judge_goal(state.goal, last_response)
-        state.last_verdict = verdict
-        state.last_reason = reason
-
-        if verdict == "done":
-            state.status = "done"
-            save_goal(self.session_id, state)
-            return {
-                "status": "done",
-                "should_continue": False,
-                "continuation_prompt": None,
-                "verdict": "done",
-                "reason": reason,
-                "message": f"✓ Goal achieved: {reason}",
-            }
-
-        if state.turns_used >= state.max_turns:
-            state.status = "paused"
-            state.paused_reason = f"turn budget exhausted ({state.turns_used}/{state.max_turns})"
-            save_goal(self.session_id, state)
-            return {
-                "status": "paused",
-                "should_continue": False,
-                "continuation_prompt": None,
-                "verdict": "continue",
-                "reason": reason,
-                "message": (
-                    f"⏸ Goal paused — {state.turns_used}/{state.max_turns} turns used. "
-                    "Use /goal resume to keep going, or /goal clear to stop."
-                ),
-            }
-
-        save_goal(self.session_id, state)
-        return {
-            "status": "active",
-            "should_continue": True,
-            "continuation_prompt": self.next_continuation_prompt(),
-            "verdict": "continue",
-            "reason": reason,
-            "message": (
-                f"↻ Continuing toward goal ({state.turns_used}/{state.max_turns}): {reason}"
-            ),
-        }
-
-    def next_continuation_prompt(self) -> Optional[str]:
-        if not self._state or self._state.status != "active":
-            return None
-        return CONTINUATION_PROMPT_TEMPLATE.format(goal=self._state.goal)
-
-
-__all__ = [
-    "GoalState",
-    "GoalManager",
-    "CONTINUATION_PROMPT_TEMPLATE",
-    "DEFAULT_MAX_TURNS",
-    "load_goal",
-    "save_goal",
-    "clear_goal",
-    "judge_goal",
-]

hermes_cli/main.py

@@ -800,8 +800,6 @@ def _print_tui_exit_summary(session_id: Optional[str], active_session_file: Opti
 
         title = db.get_session_title(target)
         message_count = int(session.get("message_count") or 0)
-        if message_count == 0:
-            return  # No real conversation — don't show resume info
         input_tokens = int(session.get("input_tokens") or 0)
         output_tokens = int(session.get("output_tokens") or 0)
         cache_read_tokens = int(session.get("cache_read_tokens") or 0)
@@ -6675,7 +6673,6 @@ def _cmd_update_impl(args, gateway_mode: bool):
         if gateway_mode
         else None
     )
-    assume_yes = bool(getattr(args, "yes", False))
 
     print("⚕ Updating Hermes Agent...")
     print()
@@ -6795,10 +6792,8 @@ def _cmd_update_impl(args, gateway_mode: bool):
         else:
             auto_stash_ref = _stash_local_changes_if_needed(git_cmd, PROJECT_ROOT)
 
-        prompt_for_restore = (
-            auto_stash_ref is not None
-            and not assume_yes
-            and (gateway_mode or (sys.stdin.isatty() and sys.stdout.isatty()))
+        prompt_for_restore = auto_stash_ref is not None and (
+            gateway_mode or (sys.stdin.isatty() and sys.stdout.isatty())
         )
 
         # Check if there are updates
@@ -7059,10 +7054,7 @@ def _cmd_update_impl(args, gateway_mode: bool):
                 print(f"  ℹ️  {len(missing_config)} new config option(s) available")
 
             print()
-            if assume_yes:
-                print("  ℹ --yes: auto-applying config migration (skipping API-key prompts).")
-                response = "y"
-            elif gateway_mode:
+            if gateway_mode:
                 response = (
                     _gateway_prompt(
                         "Would you like to configure new options now? [Y/n]", "n"
@@ -7088,17 +7080,14 @@ def _cmd_update_impl(args, gateway_mode: bool):
 
             if response in ("", "y", "yes"):
                 print()
-                # In gateway mode OR under --yes, run auto-migrations only (no
-                # input() prompts for API keys which would hang the detached
-                # process / defeat the point of --yes).
-                results = migrate_config(
-                    interactive=not (gateway_mode or assume_yes), quiet=False
-                )
+                # In gateway mode, run auto-migrations only (no input() prompts
+                # for API keys which would hang the detached process).
+                results = migrate_config(interactive=not gateway_mode, quiet=False)
 
                 if results["env_added"] or results["config_added"]:
                     print()
                     print("✓ Configuration updated!")
-                if (gateway_mode or assume_yes) and missing_env:
+                if gateway_mode and missing_env:
                     print("  ℹ API keys require manual entry: hermes config migrate")
             else:
                 print()
@@ -7148,8 +7137,6 @@ def _cmd_update_impl(args, gateway_mode: bool):
                 supports_systemd_services,
                 _ensure_user_systemd_env,
                 find_gateway_pids,
-                find_profile_gateway_processes,
-                launch_detached_profile_gateway_restart,
                 _get_service_pids,
                 _graceful_restart_via_sigusr1,
             )
@@ -7253,7 +7240,6 @@ def _cmd_update_impl(args, gateway_mode: bool):
 
             restarted_services = []
             killed_pids = set()
-            relaunched_profiles = []
 
             # --- Systemd services (Linux) ---
             # Discover all hermes-gateway* units (default + profiles)
@@ -7443,33 +7429,7 @@ def _cmd_update_impl(args, gateway_mode: bool):
             manual_pids = find_gateway_pids(
                 exclude_pids=service_pids, all_profiles=True
             )
-            profile_processes = {
-                proc.pid: proc
-                for proc in find_profile_gateway_processes(exclude_pids=service_pids)
-                if proc.pid in manual_pids
-            }
-            for pid, proc in profile_processes.items():
-                if not launch_detached_profile_gateway_restart(proc.profile, pid):
-                    continue
-                # Prefer a graceful SIGUSR1 drain so in-flight agent runs
-                # finish before the watcher respawns the gateway.  If the
-                # gateway doesn't support SIGUSR1 or doesn't exit within
-                # the drain budget, fall back to SIGTERM — the watcher
-                # still sees the exit and relaunches either way.
-                drained = _graceful_restart_via_sigusr1(
-                    pid, drain_timeout=_drain_budget,
-                )
-                if not drained:
-                    try:
-                        os.kill(pid, _signal.SIGTERM)
-                    except (ProcessLookupError, PermissionError):
-                        pass
-                killed_pids.add(pid)
-                relaunched_profiles.append(proc.profile)
-
             for pid in manual_pids:
-                if pid in profile_processes:
-                    continue
                 try:
                     os.kill(pid, _signal.SIGTERM)
                     killed_pids.add(pid)
@@ -7480,14 +7440,11 @@ def _cmd_update_impl(args, gateway_mode: bool):
                 print()
                 for svc in restarted_services:
                     print(f"  ✓ Restarted {svc}")
-                if relaunched_profiles:
-                    names = ", ".join(relaunched_profiles)
-                    print(f"  ✓ Restarting manual gateway profile(s): {names}")
-                unmapped_count = len(killed_pids) - len(relaunched_profiles)
-                if unmapped_count:
-                    print(f"  → Stopped {unmapped_count} manual gateway process(es)")
+                if killed_pids:
+                    print(f"  → Stopped {len(killed_pids)} manual gateway process(es)")
                     print("    Restart manually: hermes gateway run")
-                    if unmapped_count > 1:
+                    # Also restart for each profile if needed
+                    if len(killed_pids) > 1:
                         print(
                             "    (or: hermes -p <profile> gateway run  for each profile)"
                         )
@@ -9904,13 +9861,6 @@ Examples:
         default=False,
         help="Force a pre-update backup for this run (off by default; overrides updates.pre_update_backup)",
     )
-    update_parser.add_argument(
-        "--yes",
-        "-y",
-        action="store_true",
-        default=False,
-        help="Assume yes for interactive prompts (config migration, stash restore). API-key entry is skipped; run 'hermes config migrate' separately for those.",
-    )
     update_parser.set_defaults(func=cmd_update)
 
     # =========================================================================

hermes_cli/model_switch.py

@@ -891,19 +891,14 @@ def switch_model(
     if not validation.get("accepted"):
         override = False
         if user_providers:
-            # user_providers is a dict: {provider_slug: config_dict}
-            for slug, cfg in user_providers.items():
-                if slug == target_provider:
-                    cfg_models = cfg.get("models", {})
-                    # Direct membership works for dict (keys) and list (strings)
-                    if new_model in cfg_models:
+            for up in user_providers:
+                if isinstance(up, dict) and up.get("provider") == target_provider:
+                    cfg_models = up.get("models", [])
+                    if new_model in cfg_models or any(
+                        m.get("name") == new_model for m in cfg_models if isinstance(m, dict)
+                    ):
                         override = True
                         break
-                    # Also accept if models is a list of dicts with 'name' field
-                    if isinstance(cfg_models, list):
-                        if any(m.get("name") == new_model for m in cfg_models if isinstance(m, dict)):
-                            override = True
-                            break
         if override:
             validation = {"accepted": True, "persist": True, "recognized": False, "message": validation.get("message", "")}
         else:
@@ -1417,17 +1412,14 @@ def list_authenticated_providers(
                         models_list = list(fb)
 
             # Prefer the endpoint's live /models list when credentials are
-            # available, unless the provider explicitly opts out via
-            # discover_models: false (e.g. dedicated endpoints that expose
-            # the entire aggregator catalog via /models).
+            # available. This keeps OpenAI-compatible relays (for example CRS)
+            # in sync when the server catalog changes without requiring the
+            # user to mirror every model into config.yaml.
             api_key = str(ep_cfg.get("api_key", "") or "").strip()
             if not api_key:
                 key_env = str(ep_cfg.get("key_env", "") or "").strip()
                 api_key = os.environ.get(key_env, "").strip() if key_env else ""
-            discover = ep_cfg.get("discover_models", True)
-            if isinstance(discover, str):
-                discover = discover.lower() not in ("false", "no", "0")
-            if api_url and api_key and discover:
+            if api_url and api_key:
                 try:
                     from hermes_cli.models import fetch_api_models
                     live_models = fetch_api_models(api_key, api_url)

hermes_cli/models.py

@@ -774,6 +774,7 @@ CANONICAL_PROVIDERS: list[ProviderEntry] = [
     ProviderEntry("nous",           "Nous Portal",              "Nous Portal (Nous Research subscription)"),
     ProviderEntry("openrouter",     "OpenRouter",               "OpenRouter (100+ models, pay-per-use)"),
     ProviderEntry("lmstudio",       "LM Studio",                "LM Studio (local desktop app with built-in model server)"),
+    ProviderEntry("ai-gateway",     "Vercel AI Gateway",        "Vercel AI Gateway (200+ models, $5 free credit, no markup)"),
     ProviderEntry("anthropic",      "Anthropic",                "Anthropic (Claude models — API key or Claude Code)"),
     ProviderEntry("openai-codex",   "OpenAI Codex",             "OpenAI Codex"),
     ProviderEntry("xiaomi",         "Xiaomi MiMo",              "Xiaomi MiMo (MiMo-V2.5 and V2 models — pro, omni, flash)"),
@@ -803,7 +804,6 @@ CANONICAL_PROVIDERS: list[ProviderEntry] = [
     ProviderEntry("opencode-go",    "OpenCode Go",              "OpenCode Go (open models, $10/month subscription)"),
     ProviderEntry("bedrock",        "AWS Bedrock",              "AWS Bedrock (Claude, Nova, Llama, DeepSeek — IAM or API key)"),
     ProviderEntry("azure-foundry",  "Azure Foundry",            "Azure Foundry (OpenAI-style or Anthropic-style endpoint — your Azure AI deployment)"),
-    ProviderEntry("ai-gateway",     "Vercel AI Gateway",        "Vercel AI Gateway"),
 ]
 
 # Derived dicts — used throughout the codebase

hermes_cli/plugins.py

@@ -33,15 +33,12 @@ so plugin-defined tools appear alongside the built-in tools.
 
 from __future__ import annotations
 
-import asyncio
 import importlib
 import importlib.metadata
 import importlib.util
-import inspect
 import logging
 import os
 import sys
-import threading
 import types
 from dataclasses import dataclass, field
 from pathlib import Path
@@ -1229,55 +1226,6 @@ def get_plugin_command_handler(name: str) -> Optional[Callable]:
     return entry["handler"] if entry else None
 
 
-_PLUGIN_COMMAND_AWAIT_TIMEOUT_SECS = 30.0
-
-
-def resolve_plugin_command_result(result: Any) -> Any:
-    """Resolve a plugin command return value, awaiting async handlers when needed.
-
-    Sync CLI/TUI dispatch sites call plugin handlers from plain functions.
-    If a handler is async, await it directly when no loop is running; if
-    we're already inside an active loop, run it in a helper thread with its
-    own loop so the caller still gets a concrete result synchronously. The
-    threaded path is bounded by a 30s timeout so a hung async handler cannot
-    wedge the terminal indefinitely.
-    """
-    if not inspect.isawaitable(result):
-        return result
-
-    try:
-        asyncio.get_running_loop()
-    except RuntimeError:
-        return asyncio.run(result)
-
-    outcome: Dict[str, Any] = {}
-    failure: Dict[str, BaseException] = {}
-    done = threading.Event()
-
-    def _runner() -> None:
-        try:
-            outcome["value"] = asyncio.run(result)
-        except BaseException as exc:  # pragma: no cover - re-raised below
-            failure["exc"] = exc
-        finally:
-            done.set()
-
-    thread = threading.Thread(
-        target=_runner,
-        name="hermes-plugin-command-await",
-        daemon=True,
-    )
-    thread.start()
-    if not done.wait(timeout=_PLUGIN_COMMAND_AWAIT_TIMEOUT_SECS):
-        raise TimeoutError(
-            "Plugin command async handler did not complete within "
-            f"{_PLUGIN_COMMAND_AWAIT_TIMEOUT_SECS:.0f}s"
-        )
-    if "exc" in failure:
-        raise failure["exc"]
-    return outcome.get("value")
-
-
 def get_plugin_commands() -> Dict[str, dict]:
     """Return the full plugin commands dict (name → {handler, description, plugin}).
 

hermes_cli/runtime_provider.py

@@ -358,20 +358,11 @@ def _get_named_custom_provider(requested_provider: str) -> Optional[Dict[str, An
         return None
     if not requested_norm.startswith("custom:"):
         try:
-            canonical = auth_mod.resolve_provider(requested_norm)
+            auth_mod.resolve_provider(requested_norm)
         except AuthError:
             pass
         else:
-            # A user-declared ``custom_providers`` entry whose name matches
-            # only an *alias* (``kimi`` → built-in ``kimi-coding``) is the
-            # user's intended target — alias rewriting would otherwise hijack
-            # the request.  We only defer to the built-in when the raw name is
-            # the canonical provider itself (``nous``, ``openrouter``, …) so
-            # accidentally shadowing a canonical provider still resolves to
-            # the built-in. See tests/hermes_cli/test_runtime_provider_resolution.py
-            # ``test_named_custom_provider_does_not_shadow_builtin_provider``.
-            if (canonical or "").strip().lower() == requested_norm:
-                return None
+            return None
 
     config = load_config()
     

hermes_cli/slack_cli.py

@@ -18,7 +18,6 @@ for reinstall when scopes/commands change.
 from __future__ import annotations
 
 import json
-import os
 import sys
 from pathlib import Path
 
@@ -129,7 +128,7 @@ def slack_manifest_command(args) -> int:
 
                 target = Path(get_hermes_home()) / "slack-manifest.json"
             except Exception:
-                target = Path(os.environ.get("HERMES_HOME") or str(Path.home() / ".hermes")) / "slack-manifest.json"
+                target = Path.home() / ".hermes" / "slack-manifest.json"
         else:
             target = Path(write_target).expanduser()
         target.parent.mkdir(parents=True, exist_ok=True)

hermes_cli/status.py

@@ -125,7 +125,6 @@ def show_status(args):
     keys = {
         "OpenRouter": "OPENROUTER_API_KEY",
         "OpenAI": "OPENAI_API_KEY",
-        "NVIDIA": "NVIDIA_API_KEY",
         "Z.AI/GLM": "GLM_API_KEY",
         "Kimi": "KIMI_API_KEY",
         "StepFun Step Plan": "STEPFUN_API_KEY",

hermes_cli/web_server.py

@@ -345,7 +345,6 @@ _CATEGORY_MERGE: Dict[str, str] = {
     "dashboard": "display",
     "code_execution": "agent",
     "prompt_caching": "agent",
-    "goals": "agent",
     # Only `telegram.reactions` currently lives under telegram — fold it in
     # with the other messaging-platform config (discord) so it isn't an
     # orphan tab of one field.

hermes_state.py

@@ -514,7 +514,7 @@ class SessionDB:
     # Session lifecycle
     # =========================================================================
 
-    def _insert_session_row(
+    def create_session(
         self,
         session_id: str,
         source: str,
@@ -523,8 +523,8 @@ class SessionDB:
         system_prompt: str = None,
         user_id: str = None,
         parent_session_id: str = None,
-    ) -> None:
-        """Shared INSERT OR IGNORE for session rows."""
+    ) -> str:
+        """Create a new session record. Returns the session_id."""
         def _do(conn):
             conn.execute(
                 """INSERT OR IGNORE INTO sessions (id, source, user_id, model, model_config,
@@ -542,11 +542,8 @@ class SessionDB:
                 ),
             )
         self._execute_write(_do)
-
-    def create_session(self, session_id: str, source: str, **kwargs) -> str:
-        """Create a new session record. Returns the session_id."""
-        self._insert_session_row(session_id, source, **kwargs)
         return session_id
+
     def end_session(self, session_id: str, end_reason: str) -> None:
         """Mark a session as ended.
 
@@ -682,41 +679,21 @@ class SessionDB:
         session_id: str,
         source: str = "unknown",
         model: str = None,
-        **kwargs,
-    ) -> str:
-        """Ensure a session row exists (INSERT OR IGNORE). Accepts optional kwargs."""
-        self._insert_session_row(session_id, source, model=model, **kwargs)
-        return session_id
-
-    def prune_empty_ghost_sessions(self, sessions_dir: "Optional[Path]" = None) -> int:
-        """Remove closed empty TUI ghost sessions (no messages, no title, >24hr old)."""
-        cutoff = time.time() - 86400  # Only sessions older than 24 hours
+    ) -> None:
+        """Ensure a session row exists, creating it with minimal metadata if absent.
 
+        Used by _flush_messages_to_session_db to recover from a failed
+        create_session() call (e.g. transient SQLite lock at agent startup).
+        INSERT OR IGNORE is safe to call even when the row already exists.
+        """
         def _do(conn):
-            rows = conn.execute("""
-                SELECT id FROM sessions
-                WHERE source = 'tui'
-                  AND title IS NULL
-                  AND ended_at IS NOT NULL
-                  AND started_at < ?
-                  AND NOT EXISTS (
-                      SELECT 1 FROM messages WHERE messages.session_id = sessions.id
-                  )
-            """, (cutoff,)).fetchall()
-            ids = [r[0] if isinstance(r, (tuple, list)) else r["id"] for r in rows]
-            if ids:
-                placeholders = ",".join("?" * len(ids))
-                conn.execute(
-                    f"DELETE FROM sessions WHERE id IN ({placeholders})", ids
-                )
-            return ids
-
-        removed_ids = self._execute_write(_do) or []
-        # Clean up any on-disk session files (belt-and-suspenders)
-        if sessions_dir and removed_ids:
-            for sid in removed_ids:
-                self._remove_session_files(sessions_dir, sid)
-        return len(removed_ids)
+            conn.execute(
+                """INSERT OR IGNORE INTO sessions
+                   (id, source, model, started_at)
+                   VALUES (?, ?, ?, ?)""",
+                (session_id, source, model, time.time()),
+            )
+        self._execute_write(_do)
 
     def get_session(self, session_id: str) -> Optional[Dict[str, Any]]:
         """Get a session by ID."""
@@ -956,7 +933,6 @@ class SessionDB:
         offset: int = 0,
         include_children: bool = False,
         project_compression_tips: bool = True,
-        order_by_last_active: bool = False,
     ) -> List[Dict[str, Any]]:
         """List sessions with preview (first user message) and last active timestamp.
 
@@ -976,14 +952,6 @@ class SessionDB:
         compressed continuations from being invisible to users while keeping
         delegate subagents and branches hidden. Pass ``False`` to return the
         raw root rows (useful for admin/debug UIs).
-
-        Pass ``order_by_last_active=True`` to sort by most-recent activity
-        instead of original conversation start time. For compression chains,
-        the "most-recent activity" is taken from the live tip (not the root),
-        so an old conversation that was compressed and continued recently
-        surfaces in the correct slot. Ordering is computed at SQL level via
-        a recursive CTE that walks compression-continuation edges, so LIMIT
-        and OFFSET still apply efficiently.
         """
         where_clauses = []
         params = []
@@ -1011,80 +979,25 @@ class SessionDB:
             params.extend(exclude_sources)
 
         where_sql = f"WHERE {' AND '.join(where_clauses)}" if where_clauses else ""
-        if order_by_last_active:
-            # Compute effective_last_active by walking each surfaced session's
-            # compression-continuation chain forward in SQL and taking the MAX
-            # timestamp across the chain. This lets us ORDER BY + LIMIT at SQL
-            # level instead of fetching every row and sorting in Python, while
-            # still surfacing old compression roots whose live tip is fresh.
-            #
-            # The CTE seeds from rows the outer WHERE admits (roots + branch
-            # children), then recursively joins forward through
-            # compression-continuation edges using the same criteria as
-            # get_compression_tip (parent.end_reason='compression' AND
-            # child.started_at >= parent.ended_at).
-            query = f"""
-                WITH RECURSIVE chain(root_id, cur_id) AS (
-                    SELECT s.id, s.id FROM sessions s {where_sql}
-                    UNION ALL
-                    SELECT c.root_id, child.id
-                    FROM chain c
-                    JOIN sessions parent ON parent.id = c.cur_id
-                    JOIN sessions child ON child.parent_session_id = c.cur_id
-                    WHERE parent.end_reason = 'compression'
-                      AND child.started_at >= parent.ended_at
-                ),
-                chain_max AS (
-                    SELECT
-                        root_id,
-                        MAX(COALESCE(
-                            (SELECT MAX(m.timestamp) FROM messages m WHERE m.session_id = cur_id),
-                            (SELECT started_at FROM sessions ss WHERE ss.id = cur_id)
-                        )) AS effective_last_active
-                    FROM chain
-                    GROUP BY root_id
-                )
-                SELECT s.*,
-                    COALESCE(
-                        (SELECT SUBSTR(REPLACE(REPLACE(m.content, X'0A', ' '), X'0D', ' '), 1, 63)
-                         FROM messages m
-                         WHERE m.session_id = s.id AND m.role = 'user' AND m.content IS NOT NULL
-                         ORDER BY m.timestamp, m.id LIMIT 1),
-                        ''
-                    ) AS _preview_raw,
-                    COALESCE(
-                        (SELECT MAX(m2.timestamp) FROM messages m2 WHERE m2.session_id = s.id),
-                        s.started_at
-                    ) AS last_active,
-                    COALESCE(cm.effective_last_active, s.started_at) AS _effective_last_active
-                FROM sessions s
-                LEFT JOIN chain_max cm ON cm.root_id = s.id
-                {where_sql}
-                ORDER BY _effective_last_active DESC, s.started_at DESC, s.id DESC
-                LIMIT ? OFFSET ?
-            """
-            # WHERE params apply twice (CTE seed + outer select).
-            params = params + params + [limit, offset]
-        else:
-            query = f"""
-                SELECT s.*,
-                    COALESCE(
-                        (SELECT SUBSTR(REPLACE(REPLACE(m.content, X'0A', ' '), X'0D', ' '), 1, 63)
-                         FROM messages m
-                         WHERE m.session_id = s.id AND m.role = 'user' AND m.content IS NOT NULL
-                         ORDER BY m.timestamp, m.id LIMIT 1),
-                        ''
-                    ) AS _preview_raw,
-                    COALESCE(
-                        (SELECT MAX(m2.timestamp) FROM messages m2 WHERE m2.session_id = s.id),
-                        s.started_at
-                    ) AS last_active
-                FROM sessions s
-                {where_sql}
-                ORDER BY s.started_at DESC
-                LIMIT ? OFFSET ?
-            """
-            params.extend([limit, offset])
+        query = f"""
+            SELECT s.*,
+                COALESCE(
+                    (SELECT SUBSTR(REPLACE(REPLACE(m.content, X'0A', ' '), X'0D', ' '), 1, 63)
+                     FROM messages m
+                     WHERE m.session_id = s.id AND m.role = 'user' AND m.content IS NOT NULL
+                     ORDER BY m.timestamp, m.id LIMIT 1),
+                    ''
+                ) AS _preview_raw,
+                COALESCE(
+                    (SELECT MAX(m2.timestamp) FROM messages m2 WHERE m2.session_id = s.id),
+                    s.started_at
+                ) AS last_active
+            FROM sessions s
+            {where_sql}
+            ORDER BY s.started_at DESC
+            LIMIT ? OFFSET ?
+        """
+        params.extend([limit, offset])
         with self._lock:
             cursor = self._conn.execute(query, params)
             rows = cursor.fetchall()
@@ -1098,8 +1011,6 @@ class SessionDB:
                 s["preview"] = text + ("..." if len(raw) > 60 else "")
             else:
                 s["preview"] = ""
-            # Drop the internal ordering column so callers see a clean dict.
-            s.pop("_effective_last_active", None)
             sessions.append(s)
 
         # Project compression roots forward to their tips. Each row whose
@@ -1177,48 +1088,6 @@ class SessionDB:
     # Message storage
     # =========================================================================
 
-    # Sentinel prefix used to distinguish JSON-encoded structured content
-    # (multimodal messages: lists of parts like text + image_url) from plain
-    # string content. The NUL byte is not legal in normal text, so this
-    # cannot collide with real user content.
-    _CONTENT_JSON_PREFIX = "\x00json:"
-
-    @classmethod
-    def _encode_content(cls, content: Any) -> Any:
-        """Serialize structured (list/dict) message content for sqlite.
-
-        sqlite3 can only bind ``str``, ``bytes``, ``int``, ``float``, and ``None``
-        to query parameters. Multimodal messages have ``content`` as a list of
-        parts (``[{"type": "text", ...}, {"type": "image_url", ...}]``), which
-        raises ``ProgrammingError: Error binding parameter N: type 'list' is
-        not supported`` when bound directly.
-
-        Returns the value unchanged when it's already a safe scalar, or a
-        sentinel-prefixed JSON string for lists/dicts. Paired with
-        :meth:`_decode_content` on read.
-        """
-        if content is None or isinstance(content, (str, bytes, int, float)):
-            return content
-        try:
-            return cls._CONTENT_JSON_PREFIX + json.dumps(content)
-        except (TypeError, ValueError):
-            # Last-resort fallback: stringify so persistence never fails.
-            return str(content)
-
-    @classmethod
-    def _decode_content(cls, content: Any) -> Any:
-        """Reverse :meth:`_encode_content`; returns scalars unchanged."""
-        if isinstance(content, str) and content.startswith(cls._CONTENT_JSON_PREFIX):
-            try:
-                return json.loads(content[len(cls._CONTENT_JSON_PREFIX):])
-            except (json.JSONDecodeError, TypeError):
-                logger.warning(
-                    "Failed to decode JSON-encoded message content; "
-                    "returning raw string"
-                )
-                return content
-        return content
-
     def append_message(
         self,
         session_id: str,
@@ -1255,9 +1124,6 @@ class SessionDB:
             if codex_message_items else None
         )
         tool_calls_json = json.dumps(tool_calls) if tool_calls else None
-        # Multimodal content (list of parts) must be JSON-encoded: sqlite3
-        # cannot bind list/dict parameters directly.
-        stored_content = self._encode_content(content)
 
         # Pre-compute tool call count
         num_tool_calls = 0
@@ -1274,7 +1140,7 @@ class SessionDB:
                 (
                     session_id,
                     role,
-                    stored_content,
+                    content,
                     tool_call_id,
                     tool_calls_json,
                     tool_name,
@@ -1357,7 +1223,7 @@ class SessionDB:
                     (
                         session_id,
                         role,
-                        self._encode_content(msg.get("content")),
+                        msg.get("content"),
                         msg.get("tool_call_id"),
                         tool_calls_json,
                         msg.get("tool_name"),
@@ -1396,8 +1262,6 @@ class SessionDB:
         result = []
         for row in rows:
             msg = dict(row)
-            if "content" in msg:
-                msg["content"] = self._decode_content(msg["content"])
             if msg.get("tool_calls"):
                 try:
                     msg["tool_calls"] = json.loads(msg["tool_calls"])
@@ -1487,15 +1351,15 @@ class SessionDB:
             placeholders = ",".join("?" for _ in session_ids)
             rows = self._conn.execute(
                 "SELECT role, content, tool_call_id, tool_calls, tool_name, "
-                "finish_reason, reasoning, reasoning_content, reasoning_details, "
-                "codex_reasoning_items, codex_message_items "
+                "reasoning, reasoning_content, reasoning_details, codex_reasoning_items, "
+                "codex_message_items "
                 f"FROM messages WHERE session_id IN ({placeholders}) ORDER BY timestamp, id",
                 tuple(session_ids),
             ).fetchall()
 
         messages = []
         for row in rows:
-            content = self._decode_content(row["content"])
+            content = row["content"]
             if row["role"] in {"user", "assistant"} and isinstance(content, str):
                 content = sanitize_context(content).strip()
             msg = {"role": row["role"], "content": content}
@@ -1513,8 +1377,6 @@ class SessionDB:
             # that replay reasoning (OpenRouter, OpenAI, Nous) receive
             # coherent multi-turn reasoning context.
             if row["role"] == "assistant":
-                if row["finish_reason"]:
-                    msg["finish_reason"] = row["finish_reason"]
                 if row["reasoning"]:
                     msg["reasoning"] = row["reasoning"]
                 if row["reasoning_content"] is not None:
@@ -1882,26 +1744,10 @@ class SessionDB:
                            )""",
                         (match["id"], match["id"]),
                     )
-                    context_msgs = []
-                    for r in ctx_cursor.fetchall():
-                        raw = r["content"]
-                        decoded = self._decode_content(raw)
-                        # Multimodal context: render a compact text-only
-                        # summary for search previews.
-                        if isinstance(decoded, list):
-                            text_parts = [
-                                p.get("text", "") for p in decoded
-                                if isinstance(p, dict) and p.get("type") == "text"
-                            ]
-                            text = " ".join(t for t in text_parts if t).strip()
-                            preview = text or "[multimodal content]"
-                        elif isinstance(decoded, str):
-                            preview = decoded
-                        else:
-                            preview = ""
-                        context_msgs.append(
-                            {"role": r["role"], "content": preview[:200]}
-                        )
+                    context_msgs = [
+                        {"role": r["role"], "content": (r["content"] or "")[:200]}
+                        for r in ctx_cursor.fetchall()
+                    ]
                 match["context"] = context_msgs
             except Exception:
                 match["context"] = []
@@ -2245,3 +2091,4 @@ class SessionDB:
             result["error"] = str(exc)
 
         return result
+

model_tools.py

@@ -356,17 +356,12 @@ def _compute_tool_definitions(
             else:
                 if not quiet_mode:
                     print(f"⚠️  Unknown toolset: {toolset_name}")
-    else:
-        # Default: start with everything
+
+    elif disabled_toolsets:
         from toolsets import get_all_toolsets
         for ts_name in get_all_toolsets():
             tools_to_include.update(resolve_toolset(ts_name))
 
-    # Always apply disabled toolsets as a subtraction step at the end.
-    # This ensures that even if a composite toolset (like hermes-cli)
-    # is enabled, any tools belonging to a disabled toolset are strictly
-    # stripped out. See issue #17309.
-    if disabled_toolsets:
         for toolset_name in disabled_toolsets:
             if validate_toolset(toolset_name):
                 resolved = resolve_toolset(toolset_name)
@@ -381,6 +376,10 @@ def _compute_tool_definitions(
             else:
                 if not quiet_mode:
                     print(f"⚠️  Unknown toolset: {toolset_name}")
+    else:
+        from toolsets import get_all_toolsets
+        for ts_name in get_all_toolsets():
+            tools_to_include.update(resolve_toolset(ts_name))
 
     # Plugin-registered tools are now resolved through the normal toolset
     # path — validate_toolset() / resolve_toolset() / get_all_toolsets()

optional-skills/migration/openclaw-migration/scripts/openclaw_to_hermes.py

@@ -2960,7 +2960,7 @@ class Migrator:
 def parse_args() -> argparse.Namespace:
     parser = argparse.ArgumentParser(description="Migrate OpenClaw user state into Hermes Agent.")
     parser.add_argument("--source", default=str(Path.home() / ".openclaw"), help="OpenClaw home directory")
-    parser.add_argument("--target", default=os.environ.get("HERMES_HOME") or str(Path.home() / ".hermes"), help="Hermes home directory")
+    parser.add_argument("--target", default=str(Path.home() / ".hermes"), help="Hermes home directory")
     parser.add_argument(
         "--workspace-target",
         help="Optional workspace root where the workspace instructions file should be copied",

optional-skills/productivity/here-now/SKILL.md

@@ -1,217 +0,0 @@
----
-name: here.now
-description: Publish static sites to {slug}.here.now and store private files in cloud Drives for agent-to-agent handoff.
-version: 1.15.3
-author: here.now
-license: MIT
-prerequisites:
-  commands: [curl, file, jq]
-platforms: [macos, linux]
-metadata:
-  hermes:
-    tags: [here.now, herenow, publish, deploy, hosting, static-site, web, share, URL, drive, storage]
-    homepage: https://here.now
-    requires_toolsets: [terminal]
----
-
-# here.now
-
-here.now lets agents publish websites and store private files in cloud Drives.
-
-Use here.now for two jobs:
-
-- **Sites**: publish websites and files at `{slug}.here.now`.
-- **Drives**: store private agent files in cloud folders.
-
-## Current docs
-
-**Before answering questions about here.now capabilities, features, or workflows, read the current docs:**
-
-→ **https://here.now/docs**
-
-Read the docs:
-
-- at the first here.now-related interaction in a conversation
-- any time the user asks how to do something
-- any time the user asks what is possible, supported, or recommended
-- before telling the user a feature is unsupported
-
-Topics that require current docs (do not rely on local skill text alone):
-
-- Drives and Drive sharing
-- custom domains
-- payments and payment gating
-- forking
-- proxy routes and service variables
-- handles and links
-- limits and quotas
-- SPA routing
-- error handling and remediation
-- feature availability
-
-**If docs and live API behavior disagree, trust the live API behavior.**
-
-If the docs fetch fails or times out, continue with the local skill and live API/script output. Prefer live API behavior for active operations.
-
-## Requirements
-
-- Required binaries: `curl`, `file`, `jq`
-- Optional environment variable: `$HERENOW_API_KEY`
-- Optional Drive token variable: `$HERENOW_DRIVE_TOKEN`
-- Optional credentials file: `~/.herenow/credentials`
-- Skill helper paths:
-  - `${HERMES_SKILL_DIR}/scripts/publish.sh` for publishing sites
-  - `${HERMES_SKILL_DIR}/scripts/drive.sh` for private Drive storage
-
-## Create a site
-
-```bash
-PUBLISH="${HERMES_SKILL_DIR}/scripts/publish.sh"
-bash "$PUBLISH" {file-or-dir} --client hermes
-```
-
-Outputs the live URL (e.g. `https://bright-canvas-a7k2.here.now/`).
-
-Under the hood this is a three-step flow: create/update -> upload files -> finalize. A site is not live until finalize succeeds.
-
-Without an API key this creates an **anonymous site** that expires in 24 hours.
-With a saved API key, the site is permanent.
-
-**File structure:** For HTML sites, place `index.html` at the root of the directory you publish, not inside a subdirectory. The directory's contents become the site root. For example, publish `my-site/` where `my-site/index.html` exists — don't publish a parent folder that contains `my-site/`.
-
-You can also publish raw files without any HTML. Single files get a rich auto-viewer (images, PDF, video, audio). Multiple files get an auto-generated directory listing with folder navigation and an image gallery.
-
-## Update an existing site
-
-```bash
-PUBLISH="${HERMES_SKILL_DIR}/scripts/publish.sh"
-bash "$PUBLISH" {file-or-dir} --slug {slug} --client hermes
-```
-
-The script auto-loads the `claimToken` from `.herenow/state.json` when updating anonymous sites. Pass `--claim-token {token}` to override.
-
-Authenticated updates require a saved API key.
-
-## Use a Drive
-
-Use a Drive when the user wants private cloud storage for agent files: documents, context, memory, plans, assets, media, research, code, and anything else that should persist without being published as a website.
-
-Every signed-in account has a default Drive named `My Drive`.
-
-```bash
-DRIVE="${HERMES_SKILL_DIR}/scripts/drive.sh"
-bash "$DRIVE" default
-bash "$DRIVE" ls "My Drive"
-bash "$DRIVE" put "My Drive" notes/today.md --from ./notes/today.md
-bash "$DRIVE" cat "My Drive" notes/today.md
-bash "$DRIVE" share "My Drive" --perms write --prefix notes/ --ttl 7d
-```
-
-Use scoped Drive tokens for agent-to-agent handoff. If you receive a `herenow_drive` share block, use its `token` as `Authorization: Bearer <token>` against `api_base`, respect `pathPrefix` when present, and preserve ETags on writes. A `pathPrefix` of `null` means full-Drive access. If the skill is available, prefer `drive.sh`; otherwise call the listed API operations directly.
-
-## API key storage
-
-The publish script reads the API key from these sources (first match wins):
-
-1. `--api-key {key}` flag (CI/scripting only — avoid in interactive use)
-2. `$HERENOW_API_KEY` environment variable
-3. `~/.herenow/credentials` file (recommended for agents)
-
-To store a key, write it to the credentials file:
-
-```bash
-mkdir -p ~/.herenow && echo "{API_KEY}" > ~/.herenow/credentials && chmod 600 ~/.herenow/credentials
-```
-
-**IMPORTANT**: After receiving an API key, save it immediately — run the command above yourself. Do not ask the user to run it manually. Avoid passing the key via CLI flags (e.g. `--api-key`) in interactive sessions; the credentials file is the preferred storage method.
-
-Never commit credentials or local state files (`~/.herenow/credentials`, `.herenow/state.json`) to source control.
-
-## Getting an API key
-
-To upgrade from anonymous (24h) to permanent sites:
-
-1. Ask the user for their email address.
-2. Request a one-time sign-in code:
-
-```bash
-curl -sS https://here.now/api/auth/agent/request-code \
-  -H "content-type: application/json" \
-  -d '{"email": "user@example.com"}'
-```
-
-3. Tell the user: "Check your inbox for a sign-in code from here.now and paste it here."
-4. Verify the code and get the API key:
-
-```bash
-curl -sS https://here.now/api/auth/agent/verify-code \
-  -H "content-type: application/json" \
-  -d '{"email":"user@example.com","code":"ABCD-2345"}'
-```
-
-5. Save the returned `apiKey` yourself (do not ask the user to do this):
-
-```bash
-mkdir -p ~/.herenow && echo "{API_KEY}" > ~/.herenow/credentials && chmod 600 ~/.herenow/credentials
-```
-
-## State file
-
-After every site create/update, the script writes to `.herenow/state.json` in the working directory:
-
-```json
-{
-  "publishes": {
-    "bright-canvas-a7k2": {
-      "siteUrl": "https://bright-canvas-a7k2.here.now/",
-      "claimToken": "abc123",
-      "claimUrl": "https://here.now/claim?slug=bright-canvas-a7k2&token=abc123",
-      "expiresAt": "2026-02-18T01:00:00.000Z"
-    }
-  }
-}
-```
-
-Before creating or updating sites, you may check this file to find prior slugs.
-Treat `.herenow/state.json` as internal cache only.
-Never present this local file path as a URL, and never use it as source of truth for auth mode, expiry, or claim URL.
-
-## What to tell the user
-
-For published sites:
-
-- Always share the `siteUrl` from the current script run.
-- Read and follow `publish_result.*` lines from script stderr to determine auth mode.
-- When `publish_result.auth_mode=authenticated`: tell the user the site is **permanent** and saved to their account. No claim URL is needed.
-- When `publish_result.auth_mode=anonymous`: tell the user the site **expires in 24 hours**. Share the claim URL (if `publish_result.claim_url` is non-empty and starts with `https://`) so they can keep it permanently. Warn that claim tokens are only returned once and cannot be recovered.
-- Never tell the user to inspect `.herenow/state.json` for claim URLs or auth status.
-
-For Drives:
-
-- Do not describe Drive files as public URLs.
-- Tell the user Drive contents are private unless shared with a scoped token.
-- When sharing access with another agent, prefer a scoped token with a narrow `pathPrefix` and short TTL.
-
-## publish.sh options
-
-| Flag                   | Description                                  |
-| ---------------------- | -------------------------------------------- |
-| `--slug {slug}`        | Update an existing site instead of creating |
-| `--claim-token {token}`| Override claim token for anonymous updates    |
-| `--title {text}`       | Viewer title (non-HTML sites)             |
-| `--description {text}` | Viewer description                            |
-| `--ttl {seconds}`      | Set expiry (authenticated only)               |
-| `--client {name}`      | Agent name for attribution (e.g. `hermes`)    |
-| `--base-url {url}`     | API base URL (default: `https://here.now`)    |
-| `--allow-nonherenow-base-url` | Allow sending auth to non-default `--base-url` |
-| `--api-key {key}`      | API key override (prefer credentials file)    |
-| `--spa`                | Enable SPA routing (serve index.html for unknown paths) |
-| `--forkable`           | Allow others to fork this site                           |
-
-## Beyond publish.sh
-
-For Drive operations, use `drive.sh` or the Drive API. For broader account and site management — delete, metadata, passwords, payments, domains, handles, links, variables, proxy routes, forking, duplication, and more — see the current docs:
-
-→ **https://here.now/docs**
-
-Full docs: https://here.now/docs

optional-skills/productivity/here-now/scripts/drive.sh

@@ -1,406 +0,0 @@
-#!/usr/bin/env bash
-set -euo pipefail
-
-BASE_URL="https://here.now"
-CREDENTIALS_FILE="$HOME/.herenow/credentials"
-API_KEY="${HERENOW_API_KEY:-}"
-DRIVE_TOKEN="${HERENOW_DRIVE_TOKEN:-}"
-ALLOW_NON_HERENOW_BASE_URL=0
-MAX_FILE_BYTES=$((500 * 1024 * 1024))
-
-usage() {
-  cat <<'USAGE'
-Usage: drive.sh [global options] <command> [args]
-
-Global options:
-  --api-key <key>        Account API key (or $HERENOW_API_KEY / ~/.herenow/credentials)
-  --token <drv_live_...> Drive token (or $HERENOW_DRIVE_TOKEN)
-  --base-url <url>       API base (default: https://here.now)
-  --allow-nonherenow-base-url
-
-Commands:
-  create [name] [--default]
-  default
-  ls
-  ls <drive> [prefix]
-  cat <drive> <path>
-  put <drive> <path> --from <local-file>
-  import <drive> <prefix> --from <local-folder> [--dry-run]
-  export <drive> <prefix> --to <local-folder> [--dry-run]
-  rm <drive> <path> [--recursive --confirm <path>]
-  share <drive> --perms read|write [--prefix notes/] [--ttl 30d] [--label text] [--manage-tokens]
-  tokens <drive>
-  revoke <drive> <tokenId>
-  delete <drive> --confirm "<drive name>"
-USAGE
-  exit 1
-}
-
-die() { echo "error: $1" >&2; exit 1; }
-
-SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
-SKILL_DIR="$(cd "${SCRIPT_DIR}/.." && pwd)"
-BUNDLED_JQ="${SKILL_DIR}/bin/jq"
-
-if [[ -x "$BUNDLED_JQ" ]]; then
-  JQ_BIN="$BUNDLED_JQ"
-elif command -v jq >/dev/null 2>&1; then
-  JQ_BIN="$(command -v jq)"
-else
-  die "requires jq"
-fi
-
-for cmd in curl file; do
-  command -v "$cmd" >/dev/null 2>&1 || die "requires $cmd"
-done
-
-while [[ $# -gt 0 ]]; do
-  case "$1" in
-    --api-key) API_KEY="$2"; shift 2 ;;
-    --token) DRIVE_TOKEN="$2"; shift 2 ;;
-    --base-url) BASE_URL="$2"; shift 2 ;;
-    --allow-nonherenow-base-url) ALLOW_NON_HERENOW_BASE_URL=1; shift ;;
-    --help|-h) usage ;;
-    --*) die "unknown global option: $1" ;;
-    *) break ;;
-  esac
-done
-
-CMD="${1:-}"
-[[ -n "$CMD" ]] || usage
-shift || true
-
-if [[ -z "$API_KEY" && -z "$DRIVE_TOKEN" && -f "$CREDENTIALS_FILE" ]]; then
-  API_KEY=$(tr -d '[:space:]' < "$CREDENTIALS_FILE")
-fi
-
-BASE_URL="${BASE_URL%/}"
-if [[ "$BASE_URL" != "https://here.now" && "$ALLOW_NON_HERENOW_BASE_URL" -ne 1 ]]; then
-  if [[ -n "$API_KEY" || -n "$DRIVE_TOKEN" ]]; then
-    die "refusing to send credentials to non-default base URL; pass --allow-nonherenow-base-url to override"
-  fi
-fi
-
-auth_header=()
-if [[ -n "$DRIVE_TOKEN" ]]; then
-  auth_header=(-H "authorization: Bearer $DRIVE_TOKEN")
-elif [[ -n "$API_KEY" ]]; then
-  auth_header=(-H "authorization: Bearer $API_KEY")
-else
-  die "missing credentials; set HERENOW_API_KEY, HERENOW_DRIVE_TOKEN, or ~/.herenow/credentials"
-fi
-
-compute_sha256() {
-  local f="$1"
-  if command -v sha256sum >/dev/null 2>&1; then
-    sha256sum "$f" | cut -d' ' -f1
-  else
-    shasum -a 256 "$f" | cut -d' ' -f1
-  fi
-}
-
-guess_content_type() {
-  local f="$1"
-  case "${f##*.}" in
-    html|htm) echo "text/html; charset=utf-8" ;;
-    css) echo "text/css; charset=utf-8" ;;
-    js|mjs) echo "text/javascript; charset=utf-8" ;;
-    json) echo "application/json; charset=utf-8" ;;
-    md|txt) echo "text/plain; charset=utf-8" ;;
-    svg) echo "image/svg+xml" ;;
-    png) echo "image/png" ;;
-    jpg|jpeg) echo "image/jpeg" ;;
-    gif) echo "image/gif" ;;
-    webp) echo "image/webp" ;;
-    pdf) echo "application/pdf" ;;
-    *) file --brief --mime-type "$f" 2>/dev/null || echo "application/octet-stream" ;;
-  esac
-}
-
-api_json() {
-  local method="$1"; shift
-  local url="$1"; shift
-  local body="${1:-}"
-  local tmp
-  tmp=$(mktemp)
-  local code
-  if [[ -n "$body" ]]; then
-    code=$(curl -sS -o "$tmp" -w "%{http_code}" -X "$method" "$url" "${auth_header[@]}" -H "content-type: application/json" -d "$body")
-  else
-    code=$(curl -sS -o "$tmp" -w "%{http_code}" -X "$method" "$url" "${auth_header[@]}")
-  fi
-  if [[ "$code" -lt 200 || "$code" -ge 300 ]]; then
-    local err
-    err=$("$JQ_BIN" -r '.error // empty' "$tmp" 2>/dev/null || true)
-    [[ -n "$err" ]] || err="$(cat "$tmp")"
-    rm -f "$tmp"
-    die "HTTP $code: $err"
-  fi
-  cat "$tmp"
-  rm -f "$tmp"
-}
-
-urlenc() {
-  "$JQ_BIN" -nr --arg v "$1" '$v|@uri'
-}
-
-urlenc_path() {
-  local path="$1"
-  local out=""
-  local part
-  IFS='/' read -r -a parts <<< "$path"
-  for part in "${parts[@]}"; do
-    [[ -n "$out" ]] && out="$out/"
-    out="$out$(urlenc "$part")"
-  done
-  echo "$out"
-}
-
-resolve_drive() {
-  local name="$1"
-  if [[ "$name" == drv_* ]]; then
-    echo "$name"
-    return
-  fi
-  if [[ -n "$DRIVE_TOKEN" ]]; then
-    die "drive tokens must reference drives by drv_ id; use account credentials to resolve drive names"
-  fi
-  if [[ "$name" == "default" || "$name" == "my-drive" || "$name" == "My Drive" ]]; then
-    api_json GET "$BASE_URL/api/v1/drives/default" | "$JQ_BIN" -r '.drive.id'
-    return
-  fi
-  local rows count
-  rows=$(api_json GET "$BASE_URL/api/v1/drives" | "$JQ_BIN" --arg n "$name" '[.drives[] | select(.name == $n)]')
-  count=$(echo "$rows" | "$JQ_BIN" 'length')
-  [[ "$count" -eq 1 ]] || die "drive name '$name' matched $count drives; use a drv_ id"
-  echo "$rows" | "$JQ_BIN" -r '.[0].id'
-}
-
-drive_head() {
-  local id="$1"
-  api_json GET "$BASE_URL/api/v1/drives/$id" | "$JQ_BIN" -r '.drive.headVersionId // .headVersionId // empty'
-}
-
-file_meta() {
-  local id="$1"
-  local path="$2"
-  local prefix
-  prefix=$(urlenc "$path")
-  api_json GET "$BASE_URL/api/v1/drives/$id/files?prefix=$prefix&limit=200" | "$JQ_BIN" -c --arg p "$path" '.files[]? | select(.path == $p)' | head -n 1
-}
-
-put_file() {
-  local drive="$1"; shift
-  local path="$1"; shift
-  local local_file=""
-  while [[ $# -gt 0 ]]; do
-    case "$1" in
-      --from) local_file="$2"; shift 2 ;;
-      *) die "unexpected put argument: $1" ;;
-    esac
-  done
-  [[ -f "$local_file" ]] || die "--from must be a file"
-  local id sz ct sha meta body upload upload_url upload_id http_code
-  id=$(resolve_drive "$drive")
-  sz=$(wc -c < "$local_file" | tr -d ' ')
-  [[ "$sz" -le "$MAX_FILE_BYTES" ]] || die "$path exceeds the $MAX_FILE_BYTES byte Drive file limit"
-  ct=$(guess_content_type "$local_file")
-  sha=$(compute_sha256 "$local_file")
-  meta=$(file_meta "$id" "$path" || true)
-  body=$("$JQ_BIN" -n --arg p "$path" --argjson s "$sz" --arg c "$ct" --arg sha "$sha" \
-    '{path:$p,size:$s,contentType:$c,sha256:$sha}')
-  if [[ -n "$meta" ]]; then
-    etag=$(echo "$meta" | "$JQ_BIN" -r '.etag')
-    body=$(echo "$body" | "$JQ_BIN" --arg e "$etag" '.ifMatch = $e')
-  else
-    body=$(echo "$body" | "$JQ_BIN" '.ifNoneMatch = "*"')
-  fi
-  upload=$(api_json POST "$BASE_URL/api/v1/drives/$id/files/uploads" "$body")
-  upload_url=$(echo "$upload" | "$JQ_BIN" -r '.uploadUrl')
-  upload_id=$(echo "$upload" | "$JQ_BIN" -r '.uploadId')
-  http_code=$(curl -sS -o /dev/null -w "%{http_code}" -X PUT "$upload_url" -H "Content-Type: $ct" --data-binary "@$local_file")
-  [[ "$http_code" -ge 200 && "$http_code" -lt 300 ]] || die "upload failed for $path (HTTP $http_code)"
-  api_json POST "$BASE_URL/api/v1/drives/$id/files/finalize" "$("$JQ_BIN" -n --arg u "$upload_id" '{uploadId:$u}')" | "$JQ_BIN" .
-}
-
-case "$CMD" in
-  create)
-    name=""
-    is_default="false"
-    while [[ $# -gt 0 ]]; do
-      case "$1" in
-        --default) is_default="true"; shift ;;
-        *) [[ -z "$name" ]] && name="$1" || die "unexpected argument: $1"; shift ;;
-      esac
-    done
-    body=$("$JQ_BIN" -n --arg n "$name" --argjson d "$is_default" '{isDefault:$d} + (if $n == "" then {} else {name:$n} end)')
-    api_json POST "$BASE_URL/api/v1/drives" "$body" | "$JQ_BIN" .
-    ;;
-  default)
-    api_json GET "$BASE_URL/api/v1/drives/default" | "$JQ_BIN" .
-    ;;
-  ls)
-    if [[ $# -eq 0 ]]; then
-      [[ -z "$DRIVE_TOKEN" ]] || die "drive tokens cannot list drives; pass a drv_ id"
-      api_json GET "$BASE_URL/api/v1/drives" | "$JQ_BIN" .
-    else
-      id=$(resolve_drive "$1")
-      prefix="${2:-}"
-      api_json GET "$BASE_URL/api/v1/drives/$id/files?prefix=$(urlenc "$prefix")" | "$JQ_BIN" .
-    fi
-    ;;
-  cat)
-    [[ $# -eq 2 ]] || die "usage: drive.sh cat <drive> <path>"
-    id=$(resolve_drive "$1")
-    curl -fsS "$BASE_URL/api/v1/drives/$id/files/$(urlenc_path "$2")" "${auth_header[@]}"
-    ;;
-  put)
-    [[ $# -ge 2 ]] || die "usage: drive.sh put <drive> <path> --from <local-file>"
-    put_file "$@"
-    ;;
-  import)
-    [[ $# -ge 2 ]] || die "usage: drive.sh import <drive> <prefix> --from <local-folder> [--dry-run]"
-    drive="$1"; prefix="${2%/}"; shift 2
-    from=""; dry=0
-    while [[ $# -gt 0 ]]; do
-      case "$1" in
-        --from) from="$2"; shift 2 ;;
-        --dry-run) dry=1; shift ;;
-        *) die "unexpected import argument: $1" ;;
-      esac
-    done
-    [[ -d "$from" ]] || die "--from must be a folder"
-    uploaded=0
-    skipped=0
-    failed=0
-    planned=0
-    while IFS= read -r -d '' f; do
-      rel="${f#$from/}"
-      [[ "$rel" == .git/* || "$rel" == node_modules/* || "$rel" == ".DS_Store" || "$rel" == */.DS_Store ]] && continue
-      planned=$((planned + 1))
-      sz=$(wc -c < "$f" | tr -d ' ')
-      if [[ "$sz" -gt "$MAX_FILE_BYTES" ]]; then
-        echo "skip oversized $f ($sz bytes > $MAX_FILE_BYTES)" >&2
-        skipped=$((skipped + 1))
-        continue
-      fi
-      dest="$rel"
-      [[ -n "$prefix" ]] && dest="$prefix/$rel"
-      if [[ "$dry" -eq 1 ]]; then
-        echo "upload $f -> $dest"
-        skipped=$((skipped + 1))
-      else
-        if (put_file "$drive" "$dest" --from "$f" >/dev/null); then
-          uploaded=$((uploaded + 1))
-        else
-          failed=$((failed + 1))
-        fi
-      fi
-    done < <(find "$from" -type f -print0 | sort -z)
-    echo "planned=$planned uploaded=$uploaded skipped=$skipped failed=$failed"
-    [[ "$failed" -eq 0 ]] || exit 1
-    ;;
-  export)
-    [[ $# -ge 2 ]] || die "usage: drive.sh export <drive> <prefix> --to <local-folder> [--dry-run]"
-    id=$(resolve_drive "$1"); prefix="${2%/}"; shift 2
-    to=""; dry=0
-    while [[ $# -gt 0 ]]; do
-      case "$1" in
-        --to) to="$2"; shift 2 ;;
-        --dry-run) dry=1; shift ;;
-        *) die "unexpected export argument: $1" ;;
-      esac
-    done
-    [[ -n "$to" ]] || die "--to is required"
-    cursor=""
-    total=0
-    while true; do
-      url="$BASE_URL/api/v1/drives/$id/files?prefix=$(urlenc "$prefix")&limit=200"
-      [[ -n "$cursor" ]] && url="$url&cursor=$(urlenc "$cursor")"
-      files=$(api_json GET "$url")
-      while IFS= read -r p; do
-        [[ -n "$p" ]] || continue
-        rel="$p"
-        [[ -n "$prefix" ]] && rel="${p#$prefix/}"
-        out="$to/$rel"
-        if [[ "$dry" -eq 1 ]]; then
-          echo "download $p -> $out"
-        else
-          mkdir -p "$(dirname "$out")"
-          curl -fsS "$BASE_URL/api/v1/drives/$id/files/$(urlenc_path "$p")" "${auth_header[@]}" -o "$out"
-        fi
-        total=$((total + 1))
-      done < <(echo "$files" | "$JQ_BIN" -r '.files[].path')
-      cursor=$(echo "$files" | "$JQ_BIN" -r '.nextCursor // empty')
-      [[ -n "$cursor" ]] || break
-    done
-    echo "files=$total"
-    ;;
-  rm)
-    [[ $# -ge 2 ]] || die "usage: drive.sh rm <drive> <path> [--recursive --confirm <path>]"
-    id=$(resolve_drive "$1"); path="$2"; shift 2
-    recursive=0; confirm=""
-    while [[ $# -gt 0 ]]; do
-      case "$1" in
-        --recursive) recursive=1; shift ;;
-        --confirm) confirm="$2"; shift 2 ;;
-        *) die "unexpected rm argument: $1" ;;
-      esac
-    done
-    if [[ "$recursive" -eq 1 ]]; then
-      [[ "$confirm" == "$path" ]] || die "recursive delete requires --confirm '$path'"
-      head=$(drive_head "$id")
-      api_json DELETE "$BASE_URL/api/v1/drives/$id/files/$(urlenc_path "$path")?recursive=true&baseVersionId=$(urlenc "$head")" | "$JQ_BIN" .
-    else
-      meta=$(file_meta "$id" "$path")
-      etag=$(echo "$meta" | "$JQ_BIN" -r '.etag')
-      curl -fsS -X DELETE "$BASE_URL/api/v1/drives/$id/files/$(urlenc_path "$path")" "${auth_header[@]}" -H "If-Match: $etag" | "$JQ_BIN" .
-    fi
-    ;;
-  share)
-    [[ $# -ge 1 ]] || die "usage: drive.sh share <drive> --perms read|write [--prefix notes/] [--ttl 30d] [--label text] [--manage-tokens]"
-    id=$(resolve_drive "$1"); shift
-    perms="write"; prefix=""; ttl=""; label=""; manage_tokens="false"
-    while [[ $# -gt 0 ]]; do
-      case "$1" in
-        --perms) perms="$2"; shift 2 ;;
-        --prefix) prefix="$2"; shift 2 ;;
-        --ttl) ttl="$2"; shift 2 ;;
-        --label) label="$2"; shift 2 ;;
-        --manage-tokens) manage_tokens="true"; shift ;;
-        *) die "unexpected share argument: $1" ;;
-      esac
-    done
-    body=$("$JQ_BIN" -n --arg p "$perms" --arg pp "$prefix" --arg ttl "$ttl" --arg label "$label" --argjson mt "$manage_tokens" \
-      '{perms:$p} + (if $mt then {manageTokens:true} else {} end) + (if $ttl == "" then {} else {ttl:$ttl} end) + (if $pp == "" then {} else {pathPrefix:$pp} end) + (if $label == "" then {} else {label:$label} end)')
-    api_json POST "$BASE_URL/api/v1/drives/$id/tokens" "$body" | "$JQ_BIN" -r '.shareBlock'
-    ;;
-  tokens)
-    [[ $# -eq 1 ]] || die "usage: drive.sh tokens <drive>"
-    id=$(resolve_drive "$1")
-    api_json GET "$BASE_URL/api/v1/drives/$id/tokens" | "$JQ_BIN" .
-    ;;
-  revoke)
-    [[ $# -eq 2 ]] || die "usage: drive.sh revoke <drive> <tokenId>"
-    id=$(resolve_drive "$1")
-    api_json DELETE "$BASE_URL/api/v1/drives/$id/tokens/$2" | "$JQ_BIN" .
-    ;;
-  delete)
-    [[ $# -ge 1 ]] || die "usage: drive.sh delete <drive> --confirm <drive name>"
-    id=$(resolve_drive "$1"); shift
-    confirm=""
-    while [[ $# -gt 0 ]]; do
-      case "$1" in
-        --confirm) confirm="$2"; shift 2 ;;
-        *) die "unexpected delete argument: $1" ;;
-      esac
-    done
-    drive=$(api_json GET "$BASE_URL/api/v1/drives/$id")
-    name=$(echo "$drive" | "$JQ_BIN" -r '.drive.name')
-    [[ "$confirm" == "$name" ]] || die "delete requires --confirm '$name'"
-    api_json DELETE "$BASE_URL/api/v1/drives/$id" | "$JQ_BIN" .
-    ;;
-  *)
-    die "unknown command: $CMD"
-    ;;
-esac

optional-skills/productivity/here-now/scripts/publish.sh

@@ -1,445 +0,0 @@
-#!/usr/bin/env bash
-set -euo pipefail
-
-BASE_URL="https://here.now"
-CREDENTIALS_FILE="$HOME/.herenow/credentials"
-API_KEY="${HERENOW_API_KEY:-}"
-API_KEY_SOURCE="none"
-if [[ -n "${HERENOW_API_KEY:-}" ]]; then
-  API_KEY_SOURCE="env"
-fi
-ALLOW_NON_HERENOW_BASE_URL=0
-SLUG=""
-CLAIM_TOKEN=""
-TITLE=""
-DESCRIPTION=""
-TTL=""
-CLIENT=""
-TARGET=""
-FORKABLE=""
-SPA_MODE=""
-FROM_DRIVE=""
-DRIVE_VERSION=""
-
-usage() {
-  cat <<'USAGE'
-Usage: publish.sh <file-or-dir> [options]
-
-Options:
-  --api-key <key>         API key (or set $HERENOW_API_KEY)
-  --slug <slug>           Update existing publish
-  --claim-token <token>   Claim token for anonymous updates
-  --title <text>          Viewer title
-  --description <text>    Viewer description
-  --ttl <seconds>         Expiry (authenticated only)
-  --client <name>         Agent name for attribution (e.g. cursor, claude-code)
-  --forkable              Allow others to fork this site
-  --spa                   Enable SPA routing
-  --from-drive <drv_...>  Publish a Drive snapshot instead of local files
-  --version <dv_...>      Drive version for --from-drive (default: current head)
-  --base-url <url>        API base (default: https://here.now)
-  --allow-nonherenow-base-url
-                         Allow auth requests to non-default API base URL
-USAGE
-  exit 1
-}
-
-die() { echo "error: $1" >&2; exit 1; }
-
-SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
-SKILL_DIR="$(cd "${SCRIPT_DIR}/.." && pwd)"
-BUNDLED_JQ="${SKILL_DIR}/bin/jq"
-
-if [[ -x "$BUNDLED_JQ" ]]; then
-  JQ_BIN="$BUNDLED_JQ"
-elif command -v jq >/dev/null 2>&1; then
-  JQ_BIN="$(command -v jq)"
-else
-  die "requires jq"
-fi
-
-for cmd in curl file; do
-  command -v "$cmd" >/dev/null 2>&1 || die "requires $cmd"
-done
-
-while [[ $# -gt 0 ]]; do
-  case "$1" in
-    --api-key)      API_KEY="$2"; API_KEY_SOURCE="flag"; shift 2 ;;
-    --slug)         SLUG="$2"; shift 2 ;;
-    --claim-token)  CLAIM_TOKEN="$2"; shift 2 ;;
-    --title)        TITLE="$2"; shift 2 ;;
-    --description)  DESCRIPTION="$2"; shift 2 ;;
-    --ttl)          TTL="$2"; shift 2 ;;
-    --client)       CLIENT="$2"; shift 2 ;;
-    --base-url)     BASE_URL="$2"; shift 2 ;;
-    --allow-nonherenow-base-url) ALLOW_NON_HERENOW_BASE_URL=1; shift ;;
-    --forkable)     FORKABLE="true"; shift ;;
-    --spa)          SPA_MODE="true"; shift ;;
-    --from-drive)   FROM_DRIVE="$2"; shift 2 ;;
-    --version)      DRIVE_VERSION="$2"; shift 2 ;;
-    --help|-h)      usage ;;
-    -*)             die "unknown option: $1" ;;
-    *)              [[ -z "$TARGET" ]] && TARGET="$1" || die "unexpected argument: $1"; shift ;;
-  esac
-done
-
-if [[ -n "$FROM_DRIVE" ]]; then
-  [[ -z "$TARGET" ]] || die "--from-drive does not accept a local file-or-dir argument"
-else
-  [[ -n "$TARGET" ]] || usage
-  [[ -e "$TARGET" ]] || die "path does not exist: $TARGET"
-fi
-
-# Load API key from credentials file if not provided via flag or env
-if [[ -z "$API_KEY" && -f "$CREDENTIALS_FILE" ]]; then
-  API_KEY=$(cat "$CREDENTIALS_FILE" | tr -d '[:space:]')
-  [[ -n "$API_KEY" ]] && API_KEY_SOURCE="credentials"
-fi
-
-BASE_URL="${BASE_URL%/}"
-STATE_DIR=".herenow"
-STATE_FILE="$STATE_DIR/state.json"
-
-# Safety guard: avoid accidentally sending bearer auth to arbitrary endpoints.
-if [[ -n "$API_KEY" && "$BASE_URL" != "https://here.now" && "$ALLOW_NON_HERENOW_BASE_URL" -ne 1 ]]; then
-  die "refusing to send API key to non-default base URL; pass --allow-nonherenow-base-url to override"
-fi
-
-# Auto-load claim token from state file for anonymous updates
-if [[ -n "$SLUG" && -z "$CLAIM_TOKEN" && -z "$API_KEY" && -f "$STATE_FILE" ]]; then
-  CLAIM_TOKEN=$("$JQ_BIN" -r --arg s "$SLUG" '.publishes[$s].claimToken // empty' "$STATE_FILE" 2>/dev/null || true)
-fi
-
-if [[ -n "$FROM_DRIVE" ]]; then
-  [[ -n "$API_KEY" ]] || die "--from-drive requires an account API key"
-  BODY=$("$JQ_BIN" -n --arg d "$FROM_DRIVE" '{driveId:$d}')
-  [[ -n "$DRIVE_VERSION" ]] && BODY=$(echo "$BODY" | "$JQ_BIN" --arg v "$DRIVE_VERSION" '.versionId = $v')
-  [[ -n "$SLUG" ]] && BODY=$(echo "$BODY" | "$JQ_BIN" --arg s "$SLUG" '.slug = $s')
-  if [[ -n "$TITLE" || -n "$DESCRIPTION" ]]; then
-    viewer="{}"
-    [[ -n "$TITLE" ]] && viewer=$(echo "$viewer" | "$JQ_BIN" --arg t "$TITLE" '.title = $t')
-    [[ -n "$DESCRIPTION" ]] && viewer=$(echo "$viewer" | "$JQ_BIN" --arg d "$DESCRIPTION" '.description = $d')
-    BODY=$(echo "$BODY" | "$JQ_BIN" --argjson v "$viewer" '.viewer = $v')
-  fi
-  [[ "$FORKABLE" == "true" ]] && BODY=$(echo "$BODY" | "$JQ_BIN" '.forkable = true')
-  [[ "$SPA_MODE" == "true" ]] && BODY=$(echo "$BODY" | "$JQ_BIN" '.spaMode = true')
-  CLIENT_HEADER_VALUE="here-now-publish-sh"
-  if [[ -n "$CLIENT" ]]; then
-    normalized_client=$(echo "$CLIENT" | tr '[:upper:]' '[:lower:]' | tr -cs 'a-z0-9._-' '-')
-    normalized_client="${normalized_client#-}"
-    normalized_client="${normalized_client%-}"
-    if [[ -n "$normalized_client" ]]; then
-      CLIENT_HEADER_VALUE="${normalized_client}/publish-sh"
-    fi
-  fi
-
-  echo "publishing from Drive..." >&2
-  RESPONSE=$(curl -sS -X POST "$BASE_URL/api/v1/publish/from-drive" \
-    -H "authorization: Bearer $API_KEY" \
-    -H "x-herenow-client: $CLIENT_HEADER_VALUE" \
-    -H "content-type: application/json" \
-    -d "$BODY")
-  if echo "$RESPONSE" | "$JQ_BIN" -e '.error' >/dev/null 2>&1; then
-    err=$(echo "$RESPONSE" | "$JQ_BIN" -r '.error')
-    die "$err"
-  fi
-  SITE_URL=$(echo "$RESPONSE" | "$JQ_BIN" -r '.siteUrl')
-  OUT_SLUG=$(echo "$RESPONSE" | "$JQ_BIN" -r '.slug')
-  CURRENT_VERSION=$(echo "$RESPONSE" | "$JQ_BIN" -r '.currentVersionId')
-  DRIVE_VERSION_OUT=$(echo "$RESPONSE" | "$JQ_BIN" -r '.driveVersionId')
-  echo "$SITE_URL"
-  echo "" >&2
-  echo "publish_result.site_url=$SITE_URL" >&2
-  echo "publish_result.slug=$OUT_SLUG" >&2
-  echo "publish_result.action=from_drive" >&2
-  echo "publish_result.auth_mode=authenticated" >&2
-  echo "publish_result.api_key_source=$API_KEY_SOURCE" >&2
-  echo "publish_result.persistence=permanent" >&2
-  echo "publish_result.drive_id=$FROM_DRIVE" >&2
-  echo "publish_result.drive_version_id=$DRIVE_VERSION_OUT" >&2
-  echo "publish_result.current_version_id=$CURRENT_VERSION" >&2
-  exit 0
-fi
-
-compute_sha256() {
-  local f="$1"
-  if command -v sha256sum >/dev/null 2>&1; then
-    sha256sum "$f" | cut -d' ' -f1
-  else
-    shasum -a 256 "$f" | cut -d' ' -f1
-  fi
-}
-
-guess_content_type() {
-  local f="$1"
-  case "${f##*.}" in
-    html|htm) echo "text/html; charset=utf-8" ;;
-    css)      echo "text/css; charset=utf-8" ;;
-    js|mjs)   echo "text/javascript; charset=utf-8" ;;
-    json)     echo "application/json; charset=utf-8" ;;
-    md|txt)   echo "text/plain; charset=utf-8" ;;
-    svg)      echo "image/svg+xml" ;;
-    png)      echo "image/png" ;;
-    jpg|jpeg) echo "image/jpeg" ;;
-    gif)      echo "image/gif" ;;
-    webp)     echo "image/webp" ;;
-    pdf)      echo "application/pdf" ;;
-    mp4)      echo "video/mp4" ;;
-    mov)      echo "video/quicktime" ;;
-    mp3)      echo "audio/mpeg" ;;
-    wav)      echo "audio/wav" ;;
-    xml)      echo "application/xml" ;;
-    woff2)    echo "font/woff2" ;;
-    woff)     echo "font/woff" ;;
-    ttf)      echo "font/ttf" ;;
-    ico)      echo "image/x-icon" ;;
-    *)
-      local detected
-      detected=$(file --brief --mime-type "$f" 2>/dev/null || echo "application/octet-stream")
-      echo "$detected"
-      ;;
-  esac
-}
-
-# Build file manifest as JSON array
-FILES_JSON="[]"
-
-if [[ -f "$TARGET" ]]; then
-  sz=$(wc -c < "$TARGET" | tr -d ' ')
-  ct=$(guess_content_type "$TARGET")
-  bn=$(basename "$TARGET")
-  h=$(compute_sha256 "$TARGET")
-  FILES_JSON=$("$JQ_BIN" -n --arg p "$bn" --argjson s "$sz" --arg c "$ct" --arg h "$h" \
-    '[{"path":$p,"size":$s,"contentType":$c,"hash":$h}]')
-  FILE_MAP=$("$JQ_BIN" -n --arg p "$bn" --arg a "$(cd "$(dirname "$TARGET")" && pwd)/$(basename "$TARGET")" \
-    '{($p):$a}')
-elif [[ -d "$TARGET" ]]; then
-  FILE_MAP="{}"
-  while IFS= read -r -d '' f; do
-    rel="${f#$TARGET/}"
-    [[ "$rel" == ".DS_Store" ]] && continue
-    [[ "$(basename "$rel")" == ".DS_Store" ]] && continue
-    [[ "$rel" == ".herenow/fork-meta.json" ]] && continue
-    sz=$(wc -c < "$f" | tr -d ' ')
-    ct=$(guess_content_type "$f")
-    h=$(compute_sha256 "$f")
-    abs=$(cd "$(dirname "$f")" && pwd)/$(basename "$f")
-    FILES_JSON=$(echo "$FILES_JSON" | "$JQ_BIN" --arg p "$rel" --argjson s "$sz" --arg c "$ct" --arg h "$h" \
-      '. + [{"path":$p,"size":$s,"contentType":$c,"hash":$h}]')
-    FILE_MAP=$(echo "$FILE_MAP" | "$JQ_BIN" --arg p "$rel" --arg a "$abs" '. + {($p):$a}')
-  done < <(find "$TARGET" -type f -print0 | sort -z)
-else
-  die "not a file or directory: $TARGET"
-fi
-
-file_count=$(echo "$FILES_JSON" | "$JQ_BIN" 'length')
-[[ "$file_count" -gt 0 ]] || die "no files found"
-
-# Read fork-meta.json defaults if present and no explicit flags given
-FORK_META=""
-if [[ -d "$TARGET" ]]; then
-  FORK_META_PATH="$TARGET/.herenow/fork-meta.json"
-  if [[ -f "$FORK_META_PATH" ]]; then
-    FORK_META=$(cat "$FORK_META_PATH")
-    if [[ -z "$FORKABLE" ]]; then
-      FORKABLE=$("$JQ_BIN" -r '.forkable // empty' <<< "$FORK_META" 2>/dev/null || true)
-    fi
-  fi
-fi
-
-# Build request body
-BODY=$(echo "$FILES_JSON" | "$JQ_BIN" '{files: .}')
-
-if [[ -n "$TTL" ]]; then
-  BODY=$(echo "$BODY" | "$JQ_BIN" --argjson t "$TTL" '.ttlSeconds = $t')
-fi
-
-if [[ -n "$TITLE" || -n "$DESCRIPTION" ]]; then
-  viewer="{}"
-  [[ -n "$TITLE" ]] && viewer=$(echo "$viewer" | "$JQ_BIN" --arg t "$TITLE" '.title = $t')
-  [[ -n "$DESCRIPTION" ]] && viewer=$(echo "$viewer" | "$JQ_BIN" --arg d "$DESCRIPTION" '.description = $d')
-  BODY=$(echo "$BODY" | "$JQ_BIN" --argjson v "$viewer" '.viewer = $v')
-fi
-
-if [[ -n "$CLAIM_TOKEN" && -n "$SLUG" && -z "$API_KEY" ]]; then
-  BODY=$(echo "$BODY" | "$JQ_BIN" --arg ct "$CLAIM_TOKEN" '.claimToken = $ct')
-fi
-
-if [[ "$FORKABLE" == "true" ]]; then
-  BODY=$(echo "$BODY" | "$JQ_BIN" '.forkable = true')
-fi
-
-if [[ "$SPA_MODE" == "true" ]]; then
-  BODY=$(echo "$BODY" | "$JQ_BIN" '.spaMode = true')
-fi
-
-# Determine endpoint and method
-if [[ -n "$SLUG" ]]; then
-  URL="$BASE_URL/api/v1/publish/$SLUG"
-  METHOD="PUT"
-else
-  URL="$BASE_URL/api/v1/publish"
-  METHOD="POST"
-fi
-
-# Build auth header
-AUTH_ARGS=()
-if [[ -n "$API_KEY" ]]; then
-  AUTH_ARGS=(-H "authorization: Bearer $API_KEY")
-fi
-
-AUTH_MODE="anonymous"
-if [[ -n "$API_KEY" ]]; then
-  AUTH_MODE="authenticated"
-fi
-
-CLIENT_HEADER_VALUE="here-now-publish-sh"
-if [[ -n "$CLIENT" ]]; then
-  normalized_client=$(echo "$CLIENT" | tr '[:upper:]' '[:lower:]' | tr -cs 'a-z0-9._-' '-')
-  normalized_client="${normalized_client#-}"
-  normalized_client="${normalized_client%-}"
-  if [[ -n "$normalized_client" ]]; then
-    CLIENT_HEADER_VALUE="${normalized_client}/publish-sh"
-  fi
-fi
-CLIENT_ARGS=(-H "x-herenow-client: $CLIENT_HEADER_VALUE")
-
-# Step 1: Create/update publish
-echo "creating publish ($file_count files)..." >&2
-RESPONSE=$(curl -sS -X "$METHOD" "$URL" \
-  "${AUTH_ARGS[@]+"${AUTH_ARGS[@]}"}" \
-  "${CLIENT_ARGS[@]+"${CLIENT_ARGS[@]}"}" \
-  -H "content-type: application/json" \
-  -d "$BODY")
-
-# Check for errors
-if echo "$RESPONSE" | "$JQ_BIN" -e '.error' >/dev/null 2>&1; then
-  err=$(echo "$RESPONSE" | "$JQ_BIN" -r '.error')
-  details=$(echo "$RESPONSE" | "$JQ_BIN" -r '.details // empty')
-  die "$err${details:+ ($details)}"
-fi
-
-OUT_SLUG=$(echo "$RESPONSE" | "$JQ_BIN" -r '.slug')
-VERSION_ID=$(echo "$RESPONSE" | "$JQ_BIN" -r '.upload.versionId')
-FINALIZE_URL=$(echo "$RESPONSE" | "$JQ_BIN" -r '.upload.finalizeUrl')
-SITE_URL=$(echo "$RESPONSE" | "$JQ_BIN" -r '.siteUrl')
-UPLOAD_COUNT=$(echo "$RESPONSE" | "$JQ_BIN" '.upload.uploads | length')
-SKIPPED_COUNT=$(echo "$RESPONSE" | "$JQ_BIN" '.upload.skipped // [] | length')
-
-[[ "$OUT_SLUG" != "null" ]] || die "unexpected response: $RESPONSE"
-
-# Step 2: Upload files (skipped files are unchanged from previous version)
-if [[ "$SKIPPED_COUNT" -gt 0 ]]; then
-  echo "uploading $UPLOAD_COUNT files ($SKIPPED_COUNT unchanged, skipped)..." >&2
-else
-  echo "uploading $UPLOAD_COUNT files..." >&2
-fi
-upload_errors=0
-
-for i in $(seq 0 $((UPLOAD_COUNT - 1))); do
-  upload_path=$(echo "$RESPONSE" | "$JQ_BIN" -r ".upload.uploads[$i].path")
-  upload_url=$(echo "$RESPONSE" | "$JQ_BIN" -r ".upload.uploads[$i].url")
-  upload_ct=$(echo "$RESPONSE" | "$JQ_BIN" -r ".upload.uploads[$i].headers[\"Content-Type\"] // empty")
-
-  if [[ -f "$TARGET" && ! -d "$TARGET" ]]; then
-    local_file="$TARGET"
-  else
-    local_file=$(echo "$FILE_MAP" | "$JQ_BIN" -r --arg p "$upload_path" '.[$p]')
-  fi
-
-  if [[ ! -f "$local_file" ]]; then
-    echo "warning: missing local file for $upload_path" >&2
-    upload_errors=$((upload_errors + 1))
-    continue
-  fi
-
-  ct_args=()
-  [[ -n "$upload_ct" ]] && ct_args=(-H "Content-Type: $upload_ct")
-
-  http_code=$(curl -sS -o /dev/null -w "%{http_code}" -X PUT "$upload_url" \
-    "${ct_args[@]+"${ct_args[@]}"}" \
-    --data-binary "@$local_file")
-
-  if [[ "$http_code" -lt 200 || "$http_code" -ge 300 ]]; then
-    echo "warning: upload failed for $upload_path (HTTP $http_code)" >&2
-    upload_errors=$((upload_errors + 1))
-  fi
-done
-
-[[ "$upload_errors" -eq 0 ]] || die "$upload_errors file(s) failed to upload"
-
-# Step 3: Finalize
-echo "finalizing..." >&2
-FIN_RESPONSE=$(curl -sS -X POST "$FINALIZE_URL" \
-  "${AUTH_ARGS[@]+"${AUTH_ARGS[@]}"}" \
-  "${CLIENT_ARGS[@]+"${CLIENT_ARGS[@]}"}" \
-  -H "content-type: application/json" \
-  -d "{\"versionId\":\"$VERSION_ID\"}")
-
-if echo "$FIN_RESPONSE" | "$JQ_BIN" -e '.error' >/dev/null 2>&1; then
-  err=$(echo "$FIN_RESPONSE" | "$JQ_BIN" -r '.error')
-  die "finalize failed: $err"
-fi
-
-# Save state
-mkdir -p "$STATE_DIR"
-if [[ -f "$STATE_FILE" ]]; then
-  STATE=$(cat "$STATE_FILE")
-else
-  STATE='{"publishes":{}}'
-fi
-
-entry=$("$JQ_BIN" -n --arg s "$SITE_URL" '{siteUrl: $s}')
-
-RESPONSE_CLAIM_TOKEN=$(echo "$RESPONSE" | "$JQ_BIN" -r '.claimToken // empty')
-RESPONSE_CLAIM_URL=$(echo "$RESPONSE" | "$JQ_BIN" -r '.claimUrl // empty')
-RESPONSE_EXPIRES=$(echo "$RESPONSE" | "$JQ_BIN" -r '.expiresAt // empty')
-
-[[ -n "$RESPONSE_CLAIM_TOKEN" ]] && entry=$(echo "$entry" | "$JQ_BIN" --arg v "$RESPONSE_CLAIM_TOKEN" '.claimToken = $v')
-[[ -n "$RESPONSE_CLAIM_URL" ]] && entry=$(echo "$entry" | "$JQ_BIN" --arg v "$RESPONSE_CLAIM_URL" '.claimUrl = $v')
-[[ -n "$RESPONSE_EXPIRES" ]] && entry=$(echo "$entry" | "$JQ_BIN" --arg v "$RESPONSE_EXPIRES" '.expiresAt = $v')
-
-STATE=$(echo "$STATE" | "$JQ_BIN" --arg slug "$OUT_SLUG" --argjson e "$entry" '.publishes[$slug] = $e')
-echo "$STATE" | "$JQ_BIN" '.' > "$STATE_FILE"
-
-# Output
-echo "$SITE_URL"
-
-PERSISTENCE="permanent"
-if [[ "$AUTH_MODE" == "anonymous" ]]; then
-  PERSISTENCE="expires_24h"
-elif [[ -n "$RESPONSE_EXPIRES" ]]; then
-  PERSISTENCE="expires_at"
-fi
-
-SAFE_CLAIM_URL=""
-if [[ -n "$RESPONSE_CLAIM_URL" && "$RESPONSE_CLAIM_URL" == https://* ]]; then
-  SAFE_CLAIM_URL="$RESPONSE_CLAIM_URL"
-fi
-
-ACTION="create"
-if [[ -n "$SLUG" ]]; then
-  ACTION="update"
-fi
-
-echo "" >&2
-echo "publish_result.site_url=$SITE_URL" >&2
-echo "publish_result.slug=$OUT_SLUG" >&2
-echo "publish_result.action=$ACTION" >&2
-echo "publish_result.auth_mode=$AUTH_MODE" >&2
-echo "publish_result.api_key_source=$API_KEY_SOURCE" >&2
-echo "publish_result.persistence=$PERSISTENCE" >&2
-echo "publish_result.expires_at=$RESPONSE_EXPIRES" >&2
-echo "publish_result.claim_url=$SAFE_CLAIM_URL" >&2
-
-if [[ "$AUTH_MODE" == "authenticated" ]]; then
-  echo "authenticated publish (permanent, saved to your account)" >&2
-else
-  echo "anonymous publish (expires in 24h)" >&2
-  if [[ -n "$SAFE_CLAIM_URL" ]]; then
-    echo "claim URL: $SAFE_CLAIM_URL" >&2
-  fi
-  if [[ -n "$RESPONSE_CLAIM_TOKEN" ]]; then
-    echo "claim token saved to $STATE_FILE" >&2
-  fi
-fi

plugins/hermes-achievements/dashboard/plugin_api.py

@@ -12,14 +12,6 @@ import time
 from pathlib import Path
 from typing import Any, Dict, List, Optional, Set
 
-try:
-    from hermes_constants import get_hermes_home
-except ImportError:
-    import os as _os
-    def get_hermes_home() -> Path:  # type: ignore[misc]
-        val = (_os.environ.get("HERMES_HOME") or "").strip()
-        return Path(val) if val else Path.home() / ".hermes"
-
 try:
     from fastapi import APIRouter
 except Exception:  # Allows local unit tests without dashboard dependencies.
@@ -143,15 +135,15 @@ ACHIEVEMENTS: List[Dict[str, Any]] = [
 
 
 def state_path() -> Path:
-    return get_hermes_home() / "plugins" / "hermes-achievements" / "state.json"
+    return Path.home() / ".hermes" / "plugins" / "hermes-achievements" / "state.json"
 
 
 def snapshot_path() -> Path:
-    return get_hermes_home() / "plugins" / "hermes-achievements" / "scan_snapshot.json"
+    return Path.home() / ".hermes" / "plugins" / "hermes-achievements" / "scan_snapshot.json"
 
 
 def checkpoint_path() -> Path:
-    return get_hermes_home() / "plugins" / "hermes-achievements" / "scan_checkpoint.json"
+    return Path.home() / ".hermes" / "plugins" / "hermes-achievements" / "scan_checkpoint.json"
 
 
 def load_state() -> Dict[str, Any]:

plugins/memory/honcho/client.py

@@ -110,17 +110,6 @@ def _parse_context_tokens(host_val, root_val) -> int | None:
     return None
 
 
-def _parse_int_config(host_val, root_val, default: int) -> int:
-    """Parse an integer config: host wins, then root, then default."""
-    for val in (host_val, root_val):
-        if val is not None:
-            try:
-                return int(val)
-            except (ValueError, TypeError):
-                pass
-    return default
-
-
 def _parse_dialectic_depth(host_val, root_val) -> int:
     """Parse dialecticDepth: host wins, then root, then 1. Clamped to 1-3."""
     for val in (host_val, root_val):
@@ -474,10 +463,10 @@ class HonchoClientConfig:
                 raw.get("dialecticDynamic"),
                 default=True,
             ),
-            dialectic_max_chars=_parse_int_config(
-                host_block.get("dialecticMaxChars"),
-                raw.get("dialecticMaxChars"),
-                default=600,
+            dialectic_max_chars=int(
+                host_block.get("dialecticMaxChars")
+                or raw.get("dialecticMaxChars")
+                or 600
             ),
             dialectic_depth=_parse_dialectic_depth(
                 host_block.get("dialecticDepth"),
@@ -498,15 +487,15 @@ class HonchoClientConfig:
                 or raw.get("reasoningLevelCap")
                 or "high"
             ),
-            message_max_chars=_parse_int_config(
-                host_block.get("messageMaxChars"),
-                raw.get("messageMaxChars"),
-                default=25000,
+            message_max_chars=int(
+                host_block.get("messageMaxChars")
+                or raw.get("messageMaxChars")
+                or 25000
             ),
-            dialectic_max_input_chars=_parse_int_config(
-                host_block.get("dialecticMaxInputChars"),
-                raw.get("dialecticMaxInputChars"),
-                default=10000,
+            dialectic_max_input_chars=int(
+                host_block.get("dialecticMaxInputChars")
+                or raw.get("dialecticMaxInputChars")
+                or 10000
             ),
             recall_mode=_normalize_recall_mode(
                 host_block.get("recallMode")

plugins/memory/honcho/session.py

@@ -160,13 +160,11 @@ class HonchoSessionManager:
         Peers are lazy -- no API call until first use.
         Observation settings are controlled per-session via SessionPeerConfig.
         """
-        with self._cache_lock:
-            if peer_id in self._peers_cache:
-                return self._peers_cache[peer_id]
+        if peer_id in self._peers_cache:
+            return self._peers_cache[peer_id]
 
         peer = self.honcho.peer(peer_id)
-        with self._cache_lock:
-            self._peers_cache[peer_id] = peer
+        self._peers_cache[peer_id] = peer
         return peer
 
     def _get_or_create_honcho_session(
@@ -178,10 +176,9 @@ class HonchoSessionManager:
         Returns:
             Tuple of (honcho_session, existing_messages).
         """
-        with self._cache_lock:
-            if session_id in self._sessions_cache:
-                logger.debug("Honcho session '%s' retrieved from cache", session_id)
-                return self._sessions_cache[session_id], []
+        if session_id in self._sessions_cache:
+            logger.debug("Honcho session '%s' retrieved from cache", session_id)
+            return self._sessions_cache[session_id], []
 
         session = self.honcho.session(session_id)
 

plugins/platforms/teams/adapter.py

@@ -38,7 +38,6 @@ except ImportError:
 
 try:
     from microsoft_teams.apps import App, ActivityContext
-    from microsoft_teams.common.http.client import ClientOptions
     from microsoft_teams.api import MessageActivity, ConversationReference
     from microsoft_teams.api.activities.typing import TypingActivityInput
     from microsoft_teams.api.activities.invoke.adaptive_card import AdaptiveCardInvokeActivity
@@ -58,7 +57,6 @@ try:
     TEAMS_SDK_AVAILABLE = True
 except ImportError:
     TEAMS_SDK_AVAILABLE = False
-    ClientOptions = None  # type: ignore[assignment,misc]
     App = None  # type: ignore[assignment,misc]
     ActivityContext = None  # type: ignore[assignment,misc]
     MessageActivity = None  # type: ignore[assignment,misc]
@@ -210,7 +208,6 @@ class TeamsAdapter(BasePlatformAdapter):
                 client_secret=self._client_secret,
                 tenant_id=self._tenant_id,
                 http_server_adapter=_AiohttpBridgeAdapter(aiohttp_app),
-                client=ClientOptions(headers={"User-Agent": "Hermes"}),
             )
 
             # Register message handler before initialize()

run_agent.py

@@ -162,13 +162,6 @@ from agent.display import (
     _detect_tool_failure,
     get_tool_emoji as _get_tool_emoji,
 )
-from agent.tool_guardrails import (
-    ToolCallGuardrailConfig,
-    ToolCallGuardrailController,
-    ToolGuardrailDecision,
-    append_toolguard_guidance,
-    toolguard_synthetic_result,
-)
 from agent.trajectory import (
     convert_scratchpad_to_think, has_incomplete_scratchpad,
     save_trajectory as _save_trajectory_to_file,
@@ -1157,8 +1150,6 @@ class AIAgent:
         # Tool execution state — allows _vprint during tool execution
         # even when stream consumers are registered (no tokens streaming then)
         self._executing_tools = False
-        self._tool_guardrails = ToolCallGuardrailController()
-        self._tool_guardrail_halt_decision: ToolGuardrailDecision | None = None
 
         # Interrupt mechanism for breaking out of tool loops
         self._interrupt_requested = False
@@ -1632,12 +1623,30 @@ class AIAgent:
         self._session_db = session_db
         self._parent_session_id = parent_session_id
         self._last_flushed_db_idx = 0  # tracks DB-write cursor to prevent duplicate writes
-        self._session_db_created = False  # DB row deferred to run_conversation()
-        self._session_init_model_config = {
-            "max_iterations": self.max_iterations,
-            "reasoning_config": reasoning_config,
-            "max_tokens": max_tokens,
-        }
+        if self._session_db:
+            try:
+                self._session_db.create_session(
+                    session_id=self.session_id,
+                    source=self.platform or os.environ.get("HERMES_SESSION_SOURCE", "cli"),
+                    model=self.model,
+                    model_config={
+                        "max_iterations": self.max_iterations,
+                        "reasoning_config": reasoning_config,
+                        "max_tokens": max_tokens,
+                    },
+                    user_id=None,
+                    parent_session_id=self._parent_session_id,
+                )
+            except Exception as e:
+                # Transient SQLite lock contention (e.g. CLI and gateway writing
+                # concurrently) must NOT permanently disable session_search for
+                # this agent.  Keep _session_db alive — subsequent message
+                # flushes and session_search calls will still work once the
+                # lock clears.  The session row may be missing from the index
+                # for this run, but that is recoverable (flushes upsert rows).
+                logger.warning(
+                    "Session DB create_session failed (session_search still available): %s", e
+                )
         
         # In-memory todo list for task planning (one per agent/session)
         from tools.todo_tool import TodoStore
@@ -1649,14 +1658,6 @@ class AIAgent:
             _agent_cfg = _load_agent_config()
         except Exception:
             _agent_cfg = {}
-        try:
-            self._tool_guardrails = ToolCallGuardrailController(
-                ToolCallGuardrailConfig.from_mapping(
-                    _agent_cfg.get("tool_loop_guardrails", {})
-                )
-            )
-        except Exception as _tlg_err:
-            logger.warning("Tool loop guardrail config ignored: %s", _tlg_err)
         # Cache only the derived auxiliary compression context override that is
         # needed later by the startup feasibility check.  Avoid exposing a
         # broad pseudo-public config object on the agent instance.
@@ -2152,28 +2153,6 @@ class AIAgent:
                 "is_anthropic_oauth": self._is_anthropic_oauth,
             })
 
-    def _ensure_db_session(self) -> None:
-        """Create session DB row on first use. Disables _session_db on failure."""
-        if self._session_db_created or not self._session_db:
-            return
-        try:
-            self._session_db.create_session(
-                session_id=self.session_id,
-                source=self.platform or os.environ.get("HERMES_SESSION_SOURCE", "cli"),
-                model=self.model,
-                model_config=self._session_init_model_config,
-                system_prompt=self._cached_system_prompt,
-                user_id=None,
-                parent_session_id=self._parent_session_id,
-            )
-            self._session_db_created = True
-        except Exception as e:
-            # Transient failure (e.g. SQLite lock). Keep _session_db alive —
-            # _session_db_created stays False so next run_conversation() retries.
-            logger.warning(
-                "Session DB creation failed (will retry next turn): %s", e
-            )
-
     def reset_session_state(self):
         """Reset all session-scoped token counters to 0 for a fresh session.
         
@@ -3723,9 +3702,14 @@ class AIAgent:
             return
         self._apply_persist_user_message_override(messages)
         try:
-            # Retry row creation if the earlier attempt failed transiently.
-            if not self._session_db_created:
-                self._ensure_db_session()
+            # If create_session() failed at startup (e.g. transient lock), the
+            # session row may not exist yet.  ensure_session() uses INSERT OR
+            # IGNORE so it is a no-op when the row is already there.
+            self._session_db.ensure_session(
+                self.session_id,
+                source=self.platform or "cli",
+                model=self.model,
+            )
             start_idx = len(conversation_history) if conversation_history else 0
             flush_from = max(start_idx, self._last_flushed_db_idx)
             for msg in messages[flush_from:]:
@@ -4998,8 +4982,8 @@ class AIAgent:
     def _get_tool_call_id_static(tc) -> str:
         """Extract call ID from a tool_call entry (dict or object)."""
         if isinstance(tc, dict):
-            return tc.get("call_id", "") or tc.get("id", "") or ""
-        return getattr(tc, "call_id", "") or getattr(tc, "id", "") or ""
+            return tc.get("id", "") or ""
+        return getattr(tc, "id", "") or ""
 
     _VALID_API_ROLES = frozenset({"system", "user", "assistant", "tool", "function", "developer"})
 
@@ -8602,13 +8586,9 @@ class AIAgent:
             # message. Without it, replaying the persisted message causes
             # HTTP 400 ("The reasoning_content in the thinking mode must
             # be passed back to the API"). Include streamed reasoning
-            # text when captured; otherwise pad with a single space —
-            # DeepSeek V4 Pro tightened validation and rejects empty
-            # string ("The reasoning content in the thinking mode must
-            # be passed back to the API"). A space satisfies non-empty
-            # checks everywhere without leaking fabricated reasoning.
-            # Refs #15250, #17400, #17341.
-            msg["reasoning_content"] = reasoning_text or " "
+            # text when captured; otherwise pad with empty string.
+            # Refs #15250, #17400.
+            msg["reasoning_content"] = reasoning_text or ""
 
         # Additive fallback (refs #16844, #16884). Streaming-only providers
         # (glm, MiniMax, gpt-5.x via aigw, Anthropic via openai-compat shims)
@@ -8763,20 +8743,11 @@ class AIAgent:
             return
 
         # 1. Explicit reasoning_content already set — preserve it verbatim
-        # (includes DeepSeek/Kimi's own space-placeholder written at creation
-        # time, and any valid reasoning content from the same provider).
-        #
-        # Exception: sessions persisted BEFORE #17341 have empty-string
-        # placeholders pinned at creation time. DeepSeek V4 Pro rejects
-        # those with HTTP 400. When the active provider enforces the
-        # thinking-mode echo, upgrade "" → " " on replay so stale history
-        # doesn't 400 the user on the next turn.
+        # (includes DeepSeek/Kimi's own empty-string placeholder written at
+        # creation time, and any valid reasoning content from the same provider).
         existing = source_msg.get("reasoning_content")
         if isinstance(existing, str):
-            if existing == "" and self._needs_thinking_reasoning_pad():
-                api_msg["reasoning_content"] = " "
-            else:
-                api_msg["reasoning_content"] = existing
+            api_msg["reasoning_content"] = existing
             return
 
         needs_thinking_pad = self._needs_thinking_reasoning_pad()
@@ -8788,10 +8759,8 @@ class AIAgent:
         # pins reasoning_content at creation time for tool-call turns, so the
         # shape (reasoning set, reasoning_content absent, tool_calls present)
         # is unreachable from same-provider DeepSeek history after this fix.
-        # Inject a single space to satisfy the API without leaking another
-        # provider's chain of thought to DeepSeek/Kimi. Space (not "")
-        # because DeepSeek V4 Pro rejects empty-string reasoning_content
-        # in thinking mode (refs #17341).
+        # Inject "" to satisfy the API without leaking another provider's
+        # chain of thought to DeepSeek/Kimi.
         normalized_reasoning = source_msg.get("reasoning")
         if (
             needs_thinking_pad
@@ -8799,7 +8768,7 @@ class AIAgent:
             and isinstance(normalized_reasoning, str)
             and normalized_reasoning
         ):
-            api_msg["reasoning_content"] = " "
+            api_msg["reasoning_content"] = ""
             return
 
         # 3. Healthy session: promote 'reasoning' field to 'reasoning_content'
@@ -8812,15 +8781,12 @@ class AIAgent:
             return
 
         # 4. DeepSeek / Kimi thinking mode: all assistant messages need
-        # reasoning_content. Inject a single space to satisfy the provider's
-        # requirement when no explicit reasoning content is present. Covers
-        # both tool-call turns (already-poisoned history with no reasoning
-        # at all) and plain text turns. Space (not "") because DeepSeek V4
-        # Pro tightened validation and rejects empty string with HTTP 400
-        # ("The reasoning content in the thinking mode must be passed back
-        # to the API"). Refs #17341.
+        # reasoning_content. Inject "" to satisfy the provider's requirement
+        # when no explicit reasoning content is present. Covers both
+        # tool-call turns (already-poisoned history with no reasoning at all)
+        # and plain text turns.
         if needs_thinking_pad:
-            api_msg["reasoning_content"] = " "
+            api_msg["reasoning_content"] = ""
             return
 
         # 5. reasoning_content was present but not a string (e.g. None after
@@ -9055,15 +9021,12 @@ class AIAgent:
                 self.session_id = f"{datetime.now().strftime('%Y%m%d_%H%M%S')}_{uuid.uuid4().hex[:6]}"
                 # Update session_log_file to point to the new session's JSON file
                 self.session_log_file = self.logs_dir / f"session_{self.session_id}.json"
-                self._session_db_created = False
                 self._session_db.create_session(
                     session_id=self.session_id,
                     source=self.platform or os.environ.get("HERMES_SESSION_SOURCE", "cli"),
                     model=self.model,
-                    model_config=self._session_init_model_config,
                     parent_session_id=old_session_id,
                 )
-                self._session_db_created = True
                 # Auto-number the title for the continuation session
                 if old_title:
                     try:
@@ -9121,14 +9084,9 @@ class AIAgent:
 
         # Update token estimate after compaction so pressure calculations
         # use the post-compression count, not the stale pre-compression one.
-        # Use estimate_request_tokens_rough() so tool schemas are included —
-        # with 50+ tools enabled, schemas alone can add 20-30K tokens, and
-        # omitting them delays the next compression cycle far past the
-        # configured threshold (issue #14695).
-        _compressed_est = estimate_request_tokens_rough(
-            compressed,
-            system_prompt=new_system_prompt or "",
-            tools=self.tools or None,
+        _compressed_est = (
+            estimate_tokens_rough(new_system_prompt)
+            + estimate_messages_tokens_rough(compressed)
         )
         self.context_compressor.last_prompt_tokens = _compressed_est
         self.context_compressor.last_completion_tokens = 0
@@ -9149,44 +9107,6 @@ class AIAgent:
         )
         return compressed, new_system_prompt
 
-    def _set_tool_guardrail_halt(self, decision: ToolGuardrailDecision) -> None:
-        """Record the first guardrail decision that should stop this turn."""
-        if decision.should_halt and self._tool_guardrail_halt_decision is None:
-            self._tool_guardrail_halt_decision = decision
-
-    def _toolguard_controlled_halt_response(self, decision: ToolGuardrailDecision) -> str:
-        tool = decision.tool_name or "a tool"
-        return (
-            f"I stopped retrying {tool} because it hit the tool-call guardrail "
-            f"({decision.code}) after {decision.count} repeated non-progressing "
-            "attempts. The last tool result explains the blocker; the next step is "
-            "to change strategy instead of repeating the same call."
-        )
-
-    def _append_guardrail_observation(
-        self,
-        tool_name: str,
-        function_args: dict,
-        function_result: str,
-        *,
-        failed: bool,
-    ) -> str:
-        decision = self._tool_guardrails.after_call(
-            tool_name,
-            function_args,
-            function_result,
-            failed=failed,
-        )
-        if decision.action in {"warn", "halt"}:
-            function_result = append_toolguard_guidance(function_result, decision)
-        if decision.should_halt:
-            self._set_tool_guardrail_halt(decision)
-        return function_result
-
-    def _guardrail_block_result(self, decision: ToolGuardrailDecision) -> str:
-        self._set_tool_guardrail_halt(decision)
-        return toolguard_synthetic_result(decision)
-
     def _execute_tool_calls(self, assistant_message, messages: list, effective_task_id: str, api_call_count: int = 0) -> None:
         """Execute tool calls from the assistant message and append results to messages.
 
@@ -9230,8 +9150,7 @@ class AIAgent:
         )
 
     def _invoke_tool(self, function_name: str, function_args: dict, effective_task_id: str,
-                     tool_call_id: Optional[str] = None, messages: list = None,
-                     pre_tool_block_checked: bool = False) -> str:
+                     tool_call_id: Optional[str] = None, messages: list = None) -> str:
         """Invoke a single tool and return the result string. No display logic.
 
         Handles both agent-level tools (todo, memory, etc.) and registry-dispatched
@@ -9240,14 +9159,13 @@ class AIAgent:
         """
         # Check plugin hooks for a block directive before executing anything.
         block_message: Optional[str] = None
-        if not pre_tool_block_checked:
-            try:
-                from hermes_cli.plugins import get_pre_tool_call_block_message
-                block_message = get_pre_tool_call_block_message(
-                    function_name, function_args, task_id=effective_task_id or "",
-                )
-            except Exception:
-                pass
+        try:
+            from hermes_cli.plugins import get_pre_tool_call_block_message
+            block_message = get_pre_tool_call_block_message(
+                function_name, function_args, task_id=effective_task_id or "",
+            )
+        except Exception:
+            pass
         if block_message is not None:
             return json.dumps({"error": block_message}, ensure_ascii=False)
 
@@ -9399,31 +9317,13 @@ class AIAgent:
                 except Exception:
                     pass
 
-            block_result = None
-            blocked_by_guardrail = False
-            try:
-                from hermes_cli.plugins import get_pre_tool_call_block_message
-                block_message = get_pre_tool_call_block_message(
-                    function_name, function_args, task_id=effective_task_id or "",
-                )
-            except Exception:
-                block_message = None
-
-            if block_message is not None:
-                block_result = json.dumps({"error": block_message}, ensure_ascii=False)
-            else:
-                guardrail_decision = self._tool_guardrails.before_call(function_name, function_args)
-                if not guardrail_decision.allows_execution:
-                    block_result = self._guardrail_block_result(guardrail_decision)
-                    blocked_by_guardrail = True
-
-            parsed_calls.append((tool_call, function_name, function_args, block_result, blocked_by_guardrail))
+            parsed_calls.append((tool_call, function_name, function_args))
 
         # ── Logging / callbacks ──────────────────────────────────────────
-        tool_names_str = ", ".join(name for _, name, _, _, _ in parsed_calls)
+        tool_names_str = ", ".join(name for _, name, _ in parsed_calls)
         if not self.quiet_mode:
             print(f"  ⚡ Concurrent: {num_tools} tool calls — {tool_names_str}")
-            for i, (tc, name, args, block_result, blocked_by_guardrail) in enumerate(parsed_calls, 1):
+            for i, (tc, name, args) in enumerate(parsed_calls, 1):
                 args_str = json.dumps(args, ensure_ascii=False)
                 if self.verbose_logging:
                     print(f"  📞 Tool {i}: {name}({list(args.keys())})")
@@ -9432,9 +9332,7 @@ class AIAgent:
                     args_preview = args_str[:self.log_prefix_chars] + "..." if len(args_str) > self.log_prefix_chars else args_str
                     print(f"  📞 Tool {i}: {name}({list(args.keys())}) - {args_preview}")
 
-        for tc, name, args, block_result, blocked_by_guardrail in parsed_calls:
-            if block_result is not None:
-                continue
+        for tc, name, args in parsed_calls:
             if self.tool_progress_callback:
                 try:
                     preview = _build_tool_preview(name, args)
@@ -9442,9 +9340,7 @@ class AIAgent:
                 except Exception as cb_err:
                     logging.debug(f"Tool progress callback error: {cb_err}")
 
-        for tc, name, args, block_result, blocked_by_guardrail in parsed_calls:
-            if block_result is not None:
-                continue
+        for tc, name, args in parsed_calls:
             if self.tool_start_callback:
                 try:
                     self.tool_start_callback(tc.id, name, args)
@@ -9452,11 +9348,8 @@ class AIAgent:
                     logging.debug(f"Tool start callback error: {cb_err}")
 
         # ── Concurrent execution ─────────────────────────────────────────
-        # Each slot holds (function_name, function_args, function_result, duration, error_flag, blocked_flag)
+        # Each slot holds (function_name, function_args, function_result, duration, error_flag)
         results = [None] * num_tools
-        for i, (tc, name, args, block_result, blocked_by_guardrail) in enumerate(parsed_calls):
-            if block_result is not None:
-                results[i] = (name, args, block_result, 0.0, True, True)
 
         # Touch activity before launching workers so the gateway knows
         # we're executing tools (not stuck).
@@ -9511,14 +9404,7 @@ class AIAgent:
                     pass
             start = time.time()
             try:
-                result = self._invoke_tool(
-                    function_name,
-                    function_args,
-                    effective_task_id,
-                    tool_call.id,
-                    messages=messages,
-                    pre_tool_block_checked=True,
-                )
+                result = self._invoke_tool(function_name, function_args, effective_task_id, tool_call.id, messages=messages)
             except Exception as tool_error:
                 result = f"Error executing tool '{function_name}': {tool_error}"
                 logger.error("_invoke_tool raised for %s: %s", function_name, tool_error, exc_info=True)
@@ -9528,7 +9414,7 @@ class AIAgent:
                 logger.info("tool %s failed (%.2fs): %s", function_name, duration, result[:200])
             else:
                 logger.info("tool %s completed (%.2fs, %d chars)", function_name, duration, len(result))
-            results[index] = (function_name, function_args, result, duration, is_error, False)
+            results[index] = (function_name, function_args, result, duration, is_error)
             # Tear down worker-tid tracking.  Clear any interrupt bit we may
             # have set so the next task scheduled onto this recycled tid
             # starts with a clean slate.
@@ -9554,67 +9440,61 @@ class AIAgent:
             spinner.start()
 
         try:
-            runnable_calls = [
-                (i, tc, name, args)
-                for i, (tc, name, args, block_result, blocked_by_guardrail) in enumerate(parsed_calls)
-                if block_result is None
-            ]
-            futures = []
-            if runnable_calls:
-                max_workers = min(len(runnable_calls), _MAX_TOOL_WORKERS)
-                with concurrent.futures.ThreadPoolExecutor(max_workers=max_workers) as executor:
-                    for i, tc, name, args in runnable_calls:
-                        # Propagate ContextVars (e.g. _approval_session_key); mirrors asyncio.to_thread.
-                        ctx = contextvars.copy_context()
-                        f = executor.submit(ctx.run, _run_tool, i, tc, name, args)
-                        futures.append(f)
+            max_workers = min(num_tools, _MAX_TOOL_WORKERS)
+            with concurrent.futures.ThreadPoolExecutor(max_workers=max_workers) as executor:
+                futures = []
+                for i, (tc, name, args) in enumerate(parsed_calls):
+                    # Propagate ContextVars (e.g. _approval_session_key); mirrors asyncio.to_thread.
+                    ctx = contextvars.copy_context()
+                    f = executor.submit(ctx.run, _run_tool, i, tc, name, args)
+                    futures.append(f)
 
-                    # Wait for all to complete with periodic heartbeats so the
-                    # gateway's inactivity monitor doesn't kill us during long
-                    # concurrent tool batches. Also check for user interrupts
-                    # so we don't block indefinitely when the user sends /stop
-                    # or a new message during concurrent tool execution.
-                    _conc_start = time.time()
-                    _interrupt_logged = False
-                    while True:
-                        done, not_done = concurrent.futures.wait(
-                            futures, timeout=5.0,
-                        )
-                        if not not_done:
-                            break
+                # Wait for all to complete with periodic heartbeats so the
+                # gateway's inactivity monitor doesn't kill us during long
+                # concurrent tool batches.  Also check for user interrupts
+                # so we don't block indefinitely when the user sends /stop
+                # or a new message during concurrent tool execution.
+                _conc_start = time.time()
+                _interrupt_logged = False
+                while True:
+                    done, not_done = concurrent.futures.wait(
+                        futures, timeout=5.0,
+                    )
+                    if not not_done:
+                        break
 
-                        # Check for interrupt — the per-thread interrupt signal
-                        # already causes individual tools (terminal, execute_code)
-                        # to abort, but tools without interrupt checks (web_search,
-                        # read_file) will run to completion. Cancel any futures
-                        # that haven't started yet so we don't block on them.
-                        if self._interrupt_requested:
-                            if not _interrupt_logged:
-                                _interrupt_logged = True
-                                self._vprint(
-                                    f"{self.log_prefix}⚡ Interrupt: cancelling "
-                                    f"{len(not_done)} pending concurrent tool(s)",
-                                    force=True,
-                                )
-                            for f in not_done:
-                                f.cancel()
-                            # Give already-running tools a moment to notice the
-                            # per-thread interrupt signal and exit gracefully.
-                            concurrent.futures.wait(not_done, timeout=3.0)
-                            break
-
-                        _conc_elapsed = int(time.time() - _conc_start)
-                        # Heartbeat every ~30s (6 × 5s poll intervals)
-                        if _conc_elapsed > 0 and _conc_elapsed % 30 < 6:
-                            _still_running = [
-                                parsed_calls[futures.index(f)][1]
-                                for f in not_done
-                                if f in futures
-                            ]
-                            self._touch_activity(
-                                f"concurrent tools running ({_conc_elapsed}s, "
-                                f"{len(not_done)} remaining: {', '.join(_still_running[:3])})"
+                    # Check for interrupt — the per-thread interrupt signal
+                    # already causes individual tools (terminal, execute_code)
+                    # to abort, but tools without interrupt checks (web_search,
+                    # read_file) will run to completion.  Cancel any futures
+                    # that haven't started yet so we don't block on them.
+                    if self._interrupt_requested:
+                        if not _interrupt_logged:
+                            _interrupt_logged = True
+                            self._vprint(
+                                f"{self.log_prefix}⚡ Interrupt: cancelling "
+                                f"{len(not_done)} pending concurrent tool(s)",
+                                force=True,
                             )
+                        for f in not_done:
+                            f.cancel()
+                        # Give already-running tools a moment to notice the
+                        # per-thread interrupt signal and exit gracefully.
+                        concurrent.futures.wait(not_done, timeout=3.0)
+                        break
+
+                    _conc_elapsed = int(time.time() - _conc_start)
+                    # Heartbeat every ~30s (6 × 5s poll intervals)
+                    if _conc_elapsed > 0 and _conc_elapsed % 30 < 6:
+                        _still_running = [
+                            parsed_calls[futures.index(f)][1]
+                            for f in not_done
+                            if f in futures
+                        ]
+                        self._touch_activity(
+                            f"concurrent tools running ({_conc_elapsed}s, "
+                            f"{len(not_done)} remaining: {', '.join(_still_running[:3])})"
+                        )
         finally:
             if spinner:
                 # Build a summary message for the spinner stop
@@ -9623,9 +9503,8 @@ class AIAgent:
                 spinner.stop(f"⚡ {completed}/{num_tools} tools completed in {total_dur:.1f}s total")
 
         # ── Post-execution: display per-tool results ─────────────────────
-        for i, (tc, name, args, block_result, blocked_by_guardrail) in enumerate(parsed_calls):
+        for i, (tc, name, args) in enumerate(parsed_calls):
             r = results[i]
-            blocked = False
             if r is None:
                 # Tool was cancelled (interrupt) or thread didn't return
                 if self._interrupt_requested:
@@ -9634,21 +9513,13 @@ class AIAgent:
                     function_result = f"Error executing tool '{name}': thread did not return a result"
                 tool_duration = 0.0
             else:
-                function_name, function_args, function_result, tool_duration, is_error, blocked = r
-
-                if not blocked:
-                    function_result = self._append_guardrail_observation(
-                        function_name,
-                        function_args,
-                        function_result,
-                        failed=is_error,
-                    )
+                function_name, function_args, function_result, tool_duration, is_error = r
 
                 if is_error:
                     result_preview = function_result[:200] if len(function_result) > 200 else function_result
                     logger.warning("Tool %s returned error (%.2fs): %s", function_name, tool_duration, result_preview)
 
-                if not blocked and self.tool_progress_callback:
+                if self.tool_progress_callback:
                     try:
                         self.tool_progress_callback(
                             "tool.completed", function_name, None, None,
@@ -9676,7 +9547,7 @@ class AIAgent:
             self._current_tool = None
             self._touch_activity(f"tool completed: {name} ({tool_duration:.1f}s)")
 
-            if not blocked and self.tool_complete_callback:
+            if self.tool_complete_callback:
                 try:
                     self.tool_complete_callback(tc.id, name, args, function_result)
                 except Exception as cb_err:
@@ -9758,17 +9629,9 @@ class AIAgent:
             except Exception:
                 pass
 
-            _guardrail_block_decision: ToolGuardrailDecision | None = None
-            if _block_msg is None:
-                guardrail_decision = self._tool_guardrails.before_call(function_name, function_args)
-                if not guardrail_decision.allows_execution:
-                    _guardrail_block_decision = guardrail_decision
-
-            _execution_blocked = _block_msg is not None or _guardrail_block_decision is not None
-
-            if _execution_blocked:
-                # Tool blocked by plugin or guardrail policy — skip counters,
-                # callbacks, checkpointing, activity mutation, and real execution.
+            if _block_msg is not None:
+                # Tool blocked by plugin policy — skip counter resets.
+                # Execution is handled below in the tool dispatch chain.
                 pass
             else:
                 # Reset nudge counters when the relevant tool is actually used
@@ -9786,35 +9649,35 @@ class AIAgent:
                     args_preview = args_str[:self.log_prefix_chars] + "..." if len(args_str) > self.log_prefix_chars else args_str
                     print(f"  📞 Tool {i}: {function_name}({list(function_args.keys())}) - {args_preview}")
 
-            if not _execution_blocked:
+            if _block_msg is None:
                 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.
-            if not _execution_blocked:
+            if _block_msg is None:
                 try:
                     from tools.environments.base import set_activity_callback
                     set_activity_callback(self._touch_activity)
                 except Exception:
                     pass
 
-            if not _execution_blocked and self.tool_progress_callback:
+            if _block_msg is None and self.tool_progress_callback:
                 try:
                     preview = _build_tool_preview(function_name, function_args)
                     self.tool_progress_callback("tool.started", function_name, preview, function_args)
                 except Exception as cb_err:
                     logging.debug(f"Tool progress callback error: {cb_err}")
 
-            if not _execution_blocked and self.tool_start_callback:
+            if _block_msg is None and self.tool_start_callback:
                 try:
                     self.tool_start_callback(tool_call.id, function_name, function_args)
                 except Exception as cb_err:
                     logging.debug(f"Tool start callback error: {cb_err}")
 
             # Checkpoint: snapshot working dir before file-mutating tools
-            if not _execution_blocked and function_name in ("write_file", "patch") and self._checkpoint_mgr.enabled:
+            if _block_msg is None and function_name in ("write_file", "patch") and self._checkpoint_mgr.enabled:
                 try:
                     file_path = function_args.get("path", "")
                     if file_path:
@@ -9826,7 +9689,7 @@ class AIAgent:
                     pass  # never block tool execution
 
             # Checkpoint before destructive terminal commands
-            if not _execution_blocked and function_name == "terminal" and self._checkpoint_mgr.enabled:
+            if _block_msg is None and function_name == "terminal" and self._checkpoint_mgr.enabled:
                 try:
                     cmd = function_args.get("command", "")
                     if _is_destructive_command(cmd):
@@ -9843,11 +9706,6 @@ class AIAgent:
                 # Tool blocked by plugin policy — return error without executing.
                 function_result = json.dumps({"error": _block_msg}, ensure_ascii=False)
                 tool_duration = 0.0
-            elif _guardrail_block_decision is not None:
-                # Tool blocked by tool-loop guardrail — synthesize exactly one
-                # tool result for the original tool_call_id without executing.
-                function_result = self._guardrail_block_result(_guardrail_block_decision)
-                tool_duration = 0.0
             elif function_name == "todo":
                 from tools.todo_tool import todo_tool as _todo_tool
                 function_result = _todo_tool(
@@ -10031,22 +9889,12 @@ class AIAgent:
             # Log tool errors to the persistent error log so [error] tags
             # in the UI always have a corresponding detailed entry on disk.
             _is_error_result, _ = _detect_tool_failure(function_name, function_result)
-            if not _execution_blocked:
-                function_result = self._append_guardrail_observation(
-                    function_name,
-                    function_args,
-                    function_result,
-                    failed=_is_error_result,
-                )
-                result_preview = function_result if self.verbose_logging else (
-                    function_result[:200] if len(function_result) > 200 else function_result
-                )
             if _is_error_result:
                 logger.warning("Tool %s returned error (%.2fs): %s", function_name, tool_duration, result_preview)
             else:
                 logger.info("tool %s completed (%.2fs, %d chars)", function_name, tool_duration, len(function_result))
 
-            if not _execution_blocked and self.tool_progress_callback:
+            if self.tool_progress_callback:
                 try:
                     self.tool_progress_callback(
                         "tool.completed", function_name, None, None,
@@ -10062,7 +9910,7 @@ class AIAgent:
                 logging.debug(f"Tool {function_name} completed in {tool_duration:.2f}s")
                 logging.debug(f"Tool result ({len(function_result)} chars): {function_result}")
 
-            if not _execution_blocked and self.tool_complete_callback:
+            if self.tool_complete_callback:
                 try:
                     self.tool_complete_callback(tool_call.id, function_name, function_args, function_result)
                 except Exception as cb_err:
@@ -10165,13 +10013,6 @@ class AIAgent:
                 for idx, pfm in enumerate(self.prefill_messages):
                     api_messages.insert(sys_offset + idx, pfm.copy())
 
-            # Same safety net as the main loop: repair tool-call/result
-            # pairing before asking for a final summary.  Compression and
-            # session resume can leave a tool result whose parent assistant
-            # tool_call was summarized away; Responses API rejects that as
-            # "No tool call found for function call output".
-            api_messages = self._sanitize_api_messages(api_messages)
-
             # Same safety net as the main loop: drop thinking-only assistant
             # turns so Anthropic-family providers don't 400 the summary call.
             api_messages = self._drop_thinking_only_and_merge_users(api_messages)
@@ -10353,8 +10194,6 @@ class AIAgent:
         # Installed once, transparent when streams are healthy, prevents crash on write.
         _install_safe_stdio()
 
-        self._ensure_db_session()
-
         # 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
@@ -10398,8 +10237,6 @@ class AIAgent:
         self._last_content_tools_all_housekeeping = False
         self._mute_post_response = False
         self._unicode_sanitization_passes = 0
-        self._tool_guardrails.reset_for_turn()
-        self._tool_guardrail_halt_decision = None
 
         # Pre-turn connection health check: detect and clean up dead TCP
         # connections left over from provider outages or dropped streams.
@@ -13197,16 +13034,6 @@ class AIAgent:
 
                     self._execute_tool_calls(assistant_message, messages, effective_task_id, api_call_count)
 
-                    if self._tool_guardrail_halt_decision is not None:
-                        decision = self._tool_guardrail_halt_decision
-                        _turn_exit_reason = "guardrail_halt"
-                        final_response = self._toolguard_controlled_halt_response(decision)
-                        self._emit_status(
-                            f"⚠️ Tool guardrail halted {decision.tool_name}: {decision.code}"
-                        )
-                        messages.append({"role": "assistant", "content": final_response})
-                        break
-
                     # Reset per-turn retry counters after successful tool
                     # execution so a single truncation doesn't poison the
                     # entire conversation.
@@ -13250,13 +13077,7 @@ class AIAgent:
                         # causing premature compression.  (#12026)
                         _real_tokens = _compressor.last_prompt_tokens
                     else:
-                        # Include tool schemas — with 50+ tools enabled
-                        # these add 20-30K tokens the messages-only
-                        # estimate misses, which can skip compression
-                        # past the configured threshold (#14695).
-                        _real_tokens = estimate_request_tokens_rough(
-                            messages, tools=self.tools or None
-                        )
+                        _real_tokens = estimate_messages_tokens_rough(messages)
 
                     if self.compression_enabled and _compressor.should_compress(_real_tokens):
                         self._safe_print("  ⟳ compacting context…")
@@ -13739,7 +13560,6 @@ class AIAgent:
             "messages": messages,
             "api_calls": api_call_count,
             "completed": completed,
-            "turn_exit_reason": _turn_exit_reason,
             "partial": False,  # True only when stopped due to invalid tool calls
             "interrupted": interrupted,
             "response_previewed": getattr(self, "_response_was_previewed", False),
@@ -13759,8 +13579,6 @@ class AIAgent:
             "cost_status": self.session_cost_status,
             "cost_source": self.session_cost_source,
         }
-        if self._tool_guardrail_halt_decision is not None:
-            result["guardrail"] = self._tool_guardrail_halt_decision.to_metadata()
         # If a /steer landed after the final assistant turn (no more tool
         # batches to drain into), hand it back to the caller so it can be
         # delivered as the next user turn instead of being silently lost.

scripts/profile-tui.py

@@ -35,18 +35,10 @@ import time
 from pathlib import Path
 from typing import Any
 
-_PROJECT_ROOT = Path(__file__).resolve().parent.parent
-sys.path.insert(0, str(_PROJECT_ROOT))
-try:
-    from hermes_constants import get_hermes_home
-except ImportError:
-    def get_hermes_home() -> Path:  # type: ignore[misc]
-        val = (os.environ.get("HERMES_HOME") or "").strip()
-        return Path(val) if val else Path.home() / ".hermes"
 
 DEFAULT_TUI_DIR = Path(os.environ.get("HERMES_TUI_DIR", "/home/bb/hermes-agent/ui-tui"))
-DEFAULT_LOG = Path(os.environ.get("HERMES_PERF_LOG", str(get_hermes_home() / "perf.log")))
-DEFAULT_STATE_DB = get_hermes_home() / "state.db"
+DEFAULT_LOG = Path(os.environ.get("HERMES_PERF_LOG", str(Path.home() / ".hermes" / "perf.log")))
+DEFAULT_STATE_DB = Path.home() / ".hermes" / "state.db"
 
 # Keystroke escape sequences.  Matches what xterm/VT220 send when the
 # terminal has bracketed-paste disabled and the key-repeat handler fires.

scripts/release.py

@@ -41,17 +41,13 @@ PYPROJECT_FILE = REPO_ROOT / "pyproject.toml"
 AUTHOR_MAP = {
     # teknium (multiple emails)
     "teknium1@gmail.com": "teknium1",
-    "m@mobrienv.dev": "mikeyobrien",
     "qiyin.zuo@pcitc.com": "qiyin-code",
     "leone.parise@gmail.com": "leoneparise",
     "teknium@nousresearch.com": "teknium1",
     "127238744+teknium1@users.noreply.github.com": "teknium1",
-    "aludwin+gh@gmail.com": "adamludwin",
     "2093036+exiao@users.noreply.github.com": "exiao",
     "rylen.anil@gmail.com": "rylena",
-    "godnanijatin@gmail.com": "jatingodnani",
     "14046872+tmimmanuel@users.noreply.github.com": "tmimmanuel",
-    "657290301@qq.com": "IMHaoyan",
     "revar@users.noreply.github.com": "revaraver",
     # Matrix parity salvage batch (April 2026)
     "sr@samirusani": "samrusani",
@@ -96,8 +92,6 @@ AUTHOR_MAP = {
     "130918800+devorun@users.noreply.github.com": "devorun",
     "surat.s@itm.kmutnb.ac.th": "beesrsj2500",
     "beesr@bee.localdomain": "beesrsj2500",
-    "mind-dragon@nous.research": "Mind-Dragon",
-    "juntingpublic@gmail.com": "JustinUssuri",
     "mtf201013@gmail.com": "ma-pony",
     "sonoyuncudmr@gmail.com": "Sonoyunchu",
     "43525405+yatesjalex@users.noreply.github.com": "yatesjalex",
@@ -106,8 +100,6 @@ AUTHOR_MAP = {
     "web3blind@users.noreply.github.com": "web3blind",
     "julia@alexland.us": "alexg0bot",
     "christian@scheid.tech": "scheidti",
-    # Moonshot schema anyOf+enum salvage (May 2026)
-    "git@local.invalid": "hendrixfreire",
     "1060770+benjaminsehl@users.noreply.github.com": "benjaminsehl",
     "nerijusn76@gmail.com": "Nerijusas",
     "itonov@proton.me": "Ito-69",
@@ -120,7 +112,6 @@ AUTHOR_MAP = {
     "foxion37@gmail.com": "foxion37",
     "bloodcarter@gmail.com": "bloodcarter",
     "scott@scotttrinh.com": "scotttrinh",
-    "quocanh261997@gmail.com": "quocanh261997",
     # contributors (from noreply pattern)
     "david.vv@icloud.com": "davidvv",
     "wangqiang@wangqiangdeMac-mini.local": "xiaoqiang243",
@@ -176,7 +167,6 @@ AUTHOR_MAP = {
     "sir_even@icloud.com": "sirEven",
     "36056348+sirEven@users.noreply.github.com": "sirEven",
     "70424851+insecurejezza@users.noreply.github.com": "insecurejezza",
-    "jezzahehn@gmail.com": "JezzaHehn",
     "254021826+dodo-reach@users.noreply.github.com": "dodo-reach",
     "259807879+Bartok9@users.noreply.github.com": "Bartok9",
     "270082434+crayfish-ai@users.noreply.github.com": "crayfish-ai",
@@ -302,7 +292,6 @@ AUTHOR_MAP = {
     "154585401+LeonSGP43@users.noreply.github.com": "LeonSGP43",
     "12250313+Kailigithub@users.noreply.github.com": "Kailigithub",
     "mgparkprint@gmail.com": "vlwkaos",
-    "1317078257maroon@gmail.com": "Oxidane-bot",
     "tranquil_flow@protonmail.com": "Tranquil-Flow",
     "LyleLengyel@gmail.com": "mcndjxlefnd",
     "wangshengyang2004@163.com": "Wangshengyang2004",
@@ -341,7 +330,6 @@ AUTHOR_MAP = {
     "stefan@dimagents.ai": "dimitrovi",
     "hermes@noushq.ai": "benbarclay",
     "chinmingcock@gmail.com": "ChimingLiu",
-    "allard.quek@singtel.com": "AllardQuek",
     "openclaw@sparklab.ai": "openclaw",
     "semihcvlk53@gmail.com": "Himess",
     "erenkar950@gmail.com": "erenkarakus",
@@ -434,8 +422,6 @@ AUTHOR_MAP = {
     "ogzerber@users.noreply.github.com": "ogzerber",
     "cola-runner@users.noreply.github.com": "cola-runner",
     "ygd58@users.noreply.github.com": "ygd58",
-    "45554392+warabe1122@users.noreply.github.com": "warabe1122",
-    "187001140+willy-scr@users.noreply.github.com": "willy-scr",
     "vominh1919@users.noreply.github.com": "vominh1919",
     "iamagenius00@users.noreply.github.com": "iamagenius00",
     "9219265+cresslank@users.noreply.github.com": "cresslank",
@@ -460,7 +446,6 @@ AUTHOR_MAP = {
     "taosiyuan163@153.com": "taosiyuan163",
     "tesseracttars@gmail.com": "tesseracttars-creator",
     "tianliangjay@gmail.com": "xingkongliang",
-    "1317078257maroon@gmail.com": "Oxidane-bot",
     "tranquil_flow@protonmail.com": "Tranquil-Flow",
     "LyleLengyel@gmail.com": "mcndjxlefnd",
     "unayung@gmail.com": "Unayung",
@@ -506,11 +491,9 @@ AUTHOR_MAP = {
     "hubin_ll@qq.com": "LLQWQ",
     "memosr_email@gmail.com": "memosr",
     "jperlow@gmail.com": "perlowja",
-    "jasonpette1783@gmail.com": "web-dev0521",
     "tangyuanjc@JCdeAIfenshendeMac-mini.local": "tangyuanjc",
     "harryplusplus@gmail.com": "harryplusplus",
     "anthhub@163.com": "anthhub",
-    "allard.quek@singtel.com": "AllardQuek",
     "shenuu@gmail.com": "shenuu",
     "xiayh17@gmail.com": "xiayh0107",
     "zhujianxyz@gmail.com": "opriz",

tests/acp/test_server.py

@@ -730,7 +730,6 @@ class TestSlashCommands:
         ]
         state.agent.compression_enabled = True
         state.agent._cached_system_prompt = "system"
-        state.agent.tools = None
         original_session_db = object()
         state.agent._session_db = original_session_db
 
@@ -747,7 +746,7 @@ class TestSlashCommands:
         with (
             patch.object(agent.session_manager, "save_session") as mock_save,
             patch(
-                "agent.model_metadata.estimate_request_tokens_rough",
+                "agent.model_metadata.estimate_messages_tokens_rough",
                 side_effect=[40, 12],
             ),
         ):

tests/acp/test_session.py

@@ -8,7 +8,6 @@ from types import SimpleNamespace
 import pytest
 from unittest.mock import MagicMock, patch
 
-from acp_adapter import session as acp_session
 from acp_adapter.session import SessionManager, SessionState
 from hermes_state import SessionDB
 
@@ -43,27 +42,6 @@ class TestCreateSession:
         state = manager.create_session(cwd="/tmp/work")
         assert calls == [(state.session_id, "/tmp/work")]
 
-
-    def test_register_task_cwd_translates_windows_drive_for_wsl_tools(self, monkeypatch):
-        captured = {}
-
-        def fake_register_task_env_overrides(task_id, overrides):
-            captured["task_id"] = task_id
-            captured["overrides"] = overrides
-
-        monkeypatch.setattr("hermes_constants._wsl_detected", True)
-        monkeypatch.setattr(
-            "tools.terminal_tool.register_task_env_overrides",
-            fake_register_task_env_overrides,
-        )
-
-        acp_session._register_task_cwd("session-1", r"E:\Projects\AI\paperclip")
-
-        assert captured == {
-            "task_id": "session-1",
-            "overrides": {"cwd": "/mnt/e/Projects/AI/paperclip"},
-        }
-
     def test_session_ids_are_unique(self, manager):
         s1 = manager.create_session()
         s2 = manager.create_session()
@@ -78,59 +56,6 @@ class TestCreateSession:
         assert manager.get_session("does-not-exist") is None
 
 
-
-
-# ---------------------------------------------------------------------------
-# WSL cwd translation
-# ---------------------------------------------------------------------------
-
-
-class TestWslCwdTranslation:
-    def test_translate_acp_cwd_converts_windows_drive_path_when_wsl(self, monkeypatch):
-        monkeypatch.setattr("hermes_constants._wsl_detected", True)
-
-        assert acp_session._translate_acp_cwd(r"E:\Projects\AI\paperclip") == "/mnt/e/Projects/AI/paperclip"
-
-    def test_translate_acp_cwd_handles_forward_slashes_when_wsl(self, monkeypatch):
-        monkeypatch.setattr("hermes_constants._wsl_detected", True)
-
-        assert acp_session._translate_acp_cwd("D:/work/project") == "/mnt/d/work/project"
-
-    def test_translate_acp_cwd_leaves_windows_drive_path_unchanged_off_wsl(self, monkeypatch):
-        monkeypatch.setattr("hermes_constants._wsl_detected", False)
-
-        assert acp_session._translate_acp_cwd(r"E:\Projects\AI\paperclip") == r"E:\Projects\AI\paperclip"
-
-    def test_translate_acp_cwd_leaves_posix_path_unchanged_on_wsl(self, monkeypatch):
-        monkeypatch.setattr("hermes_constants._wsl_detected", True)
-
-        assert acp_session._translate_acp_cwd("/mnt/e/Projects/AI/paperclip") == "/mnt/e/Projects/AI/paperclip"
-
-    def test_create_session_stores_translated_cwd_on_wsl(self, manager, monkeypatch):
-        monkeypatch.setattr("hermes_constants._wsl_detected", True)
-
-        state = manager.create_session(cwd=r"E:\Projects\AI\paperclip")
-
-        assert state.cwd == "/mnt/e/Projects/AI/paperclip"
-
-    def test_fork_session_stores_translated_cwd_on_wsl(self, manager, monkeypatch):
-        monkeypatch.setattr("hermes_constants._wsl_detected", True)
-        original = manager.create_session(cwd="/tmp/base")
-
-        forked = manager.fork_session(original.session_id, cwd=r"D:\work\project")
-
-        assert forked is not None
-        assert forked.cwd == "/mnt/d/work/project"
-
-    def test_update_cwd_stores_translated_cwd_on_wsl(self, manager, monkeypatch):
-        monkeypatch.setattr("hermes_constants._wsl_detected", True)
-        state = manager.create_session(cwd="/tmp/old")
-
-        updated = manager.update_cwd(state.session_id, cwd=r"C:\Users\foo\project")
-
-        assert updated is not None
-        assert updated.cwd == "/mnt/c/Users/foo/project"
-
 # ---------------------------------------------------------------------------
 # fork
 # ---------------------------------------------------------------------------

tests/acp_adapter/test_acp_commands.py

@@ -1,150 +0,0 @@
-from types import SimpleNamespace
-
-import pytest
-from acp.schema import TextContentBlock
-
-from acp_adapter.server import HermesACPAgent
-from acp_adapter.session import SessionManager
-
-
-class FakeAgent:
-    def __init__(self):
-        self.model = "fake-model"
-        self.provider = "fake-provider"
-        self.enabled_toolsets = ["hermes-acp"]
-        self.disabled_toolsets = []
-        self.tools = []
-        self.valid_tool_names = set()
-        self.steers = []
-        self.runs = []
-
-    def steer(self, text):
-        self.steers.append(text)
-        return True
-
-    def run_conversation(self, *, user_message, conversation_history, task_id, **kwargs):
-        self.runs.append(user_message)
-        messages = list(conversation_history or [])
-        messages.append({"role": "user", "content": user_message})
-        final = f"ran: {user_message}"
-        messages.append({"role": "assistant", "content": final})
-        return {"final_response": final, "messages": messages}
-
-
-class CaptureConn:
-    def __init__(self):
-        self.updates = []
-
-    async def session_update(self, *args, **kwargs):
-        if kwargs:
-            self.updates.append((kwargs.get("session_id"), kwargs.get("update")))
-        else:
-            self.updates.append((args[0], args[1]))
-
-    async def request_permission(self, *args, **kwargs):
-        return SimpleNamespace(outcome="allow")
-
-
-class NoopDb:
-    def get_session(self, *_args, **_kwargs):
-        return None
-
-    def create_session(self, *_args, **_kwargs):
-        return None
-
-    def update_session(self, *_args, **_kwargs):
-        return None
-
-
-def make_agent_and_state():
-    fake = FakeAgent()
-    manager = SessionManager(agent_factory=lambda **kwargs: fake, db=NoopDb())
-    acp_agent = HermesACPAgent(session_manager=manager)
-    state = manager.create_session(cwd=".")
-    conn = CaptureConn()
-    acp_agent.on_connect(conn)
-    return acp_agent, state, fake, conn
-
-
-@pytest.mark.asyncio
-async def test_acp_steer_slash_command_injects_into_running_agent():
-    acp_agent, state, fake, _conn = make_agent_and_state()
-    state.is_running = True
-
-    response = await acp_agent.prompt(
-        session_id=state.session_id,
-        prompt=[TextContentBlock(type="text", text="/steer prefer the simpler fix")],
-    )
-
-    assert response.stop_reason == "end_turn"
-    assert fake.steers == ["prefer the simpler fix"]
-    assert fake.runs == []
-
-
-@pytest.mark.asyncio
-async def test_acp_steer_after_zed_interrupt_replays_interrupted_prompt_with_guidance():
-    acp_agent, state, fake, _conn = make_agent_and_state()
-    state.interrupted_prompt_text = "write hi to a text file"
-
-    response = await acp_agent.prompt(
-        session_id=state.session_id,
-        prompt=[TextContentBlock(type="text", text="/steer write HELLO instead")],
-    )
-
-    assert response.stop_reason == "end_turn"
-    assert fake.steers == []
-    assert fake.runs == [
-        "write hi to a text file\n\nUser correction/guidance after interrupt: write HELLO instead"
-    ]
-    assert state.interrupted_prompt_text == ""
-
-
-@pytest.mark.asyncio
-async def test_acp_steer_on_idle_session_runs_as_regular_prompt():
-    # /steer on an idle session (no running turn, nothing to salvage) should
-    # run the steer payload as a normal user prompt — NOT silently append it
-    # to state.queued_prompts. Without this, users on Zed / other ACP clients
-    # see their /steer turn into "queued for the next turn" when they never
-    # typed /queue. Matches gateway/run.py ~L4898 idle-/steer behavior.
-    acp_agent, state, fake, _conn = make_agent_and_state()
-
-    response = await acp_agent.prompt(
-        session_id=state.session_id,
-        prompt=[TextContentBlock(type="text", text="/steer summarize the README")],
-    )
-
-    assert response.stop_reason == "end_turn"
-    assert fake.steers == []
-    assert fake.runs == ["summarize the README"]
-    assert state.queued_prompts == []
-
-
-@pytest.mark.asyncio
-async def test_acp_queue_slash_command_adds_next_turn_without_running_now():
-    acp_agent, state, fake, _conn = make_agent_and_state()
-
-    response = await acp_agent.prompt(
-        session_id=state.session_id,
-        prompt=[TextContentBlock(type="text", text="/queue run the tests after this")],
-    )
-
-    assert response.stop_reason == "end_turn"
-    assert state.queued_prompts == ["run the tests after this"]
-    assert fake.runs == []
-
-
-@pytest.mark.asyncio
-async def test_acp_prompt_drains_queued_turns_after_current_run():
-    acp_agent, state, fake, conn = make_agent_and_state()
-    state.queued_prompts.append("then run tests")
-
-    response = await acp_agent.prompt(
-        session_id=state.session_id,
-        prompt=[TextContentBlock(type="text", text="make the change")],
-    )
-
-    assert response.stop_reason == "end_turn"
-    assert fake.runs == ["make the change", "then run tests"]
-    assert state.queued_prompts == []
-    agent_messages = [u for _sid, u in conn.updates if getattr(u, "session_update", None) == "agent_message_chunk"]
-    assert len(agent_messages) >= 2

tests/agent/test_auxiliary_named_custom_providers.py

@@ -427,68 +427,3 @@ class TestProvidersDictApiModeAnthropicMessages:
         assert isinstance(sync_client, OpenAI)
         async_client, _ = resolve_provider_client("localchat", async_mode=True)
         assert isinstance(async_client, AsyncOpenAI)
-
-
-class TestCustomProviderAliasCollision:
-    """A user-declared custom_providers entry whose name matches a built-in
-    *alias* (not a canonical provider) must win over the built-in.
-
-    Regression guard for #15743: users who defined fallback_model pointing at
-    a custom_providers entry named ``kimi`` were having requests routed to
-    the built-in kimi-coding endpoint because ``_normalize_aux_provider``
-    rewrote ``kimi`` → ``kimi-coding`` before the named-custom lookup.
-    """
-
-    def test_custom_named_kimi_wins_over_builtin_alias(self, tmp_path):
-        _write_config(tmp_path, {
-            "model": {"provider": "openrouter", "default": "anthropic/claude-sonnet-4.6"},
-            "custom_providers": [
-                {
-                    "name": "kimi",
-                    "base_url": "https://my-custom-kimi.example.com/v1",
-                    "api_key": "my-kimi-key",
-                    "models": {"my-kimi-model": {"context_length": 200000}},
-                },
-            ],
-        })
-        from agent.auxiliary_client import resolve_provider_client
-        from openai import OpenAI
-        client, model = resolve_provider_client("kimi", model="my-kimi-model", raw_codex=True)
-        assert isinstance(client, OpenAI)
-        assert "my-custom-kimi.example.com" in str(client.base_url)
-        assert client.api_key == "my-kimi-key"
-        assert model == "my-kimi-model"
-
-    def test_bare_kimi_without_custom_still_routes_to_builtin(self, tmp_path, monkeypatch):
-        """Regression guard: bare 'kimi' with no custom entry must still
-        reach the built-in kimi-coding provider."""
-        _write_config(tmp_path, {
-            "model": {"provider": "openrouter", "default": "anthropic/claude-sonnet-4.6"},
-        })
-        monkeypatch.setenv("KIMI_API_KEY", "builtin-kimi-key")
-        from agent.auxiliary_client import resolve_provider_client
-        client, _ = resolve_provider_client("kimi", model="kimi-k2-0905-preview", raw_codex=True)
-        assert client is not None
-        base_url = str(client.base_url)
-        # Built-in kimi-coding points at api.moonshot.ai
-        assert "moonshot" in base_url or "kimi" in base_url, f"unexpected base_url {base_url!r}"
-
-    def test_explicit_overrides_applied_on_api_key_branch(self, tmp_path, monkeypatch):
-        """Explicit base_url/api_key from the caller must override the
-        registered provider's defaults on the API-key branch.  Used by
-        _try_activate_fallback to route a fallback through a built-in
-        provider name but targeting a user-supplied endpoint."""
-        _write_config(tmp_path, {
-            "model": {"provider": "openrouter", "default": "anthropic/claude-sonnet-4.6"},
-        })
-        monkeypatch.setenv("KIMI_API_KEY", "builtin-kimi-key")
-        from agent.auxiliary_client import resolve_provider_client
-        from openai import OpenAI
-        client, _ = resolve_provider_client(
-            "kimi-coding", model="kimi-k2", raw_codex=True,
-            explicit_base_url="https://override.example.com",
-            explicit_api_key="override-key",
-        )
-        assert isinstance(client, OpenAI)
-        assert "override.example.com" in str(client.base_url)
-        assert client.api_key == "override-key"

tests/agent/test_context_compressor.py

@@ -640,30 +640,6 @@ class TestCompressWithClient:
                 for tc in msg["tool_calls"]:
                     assert tc["id"] in answered_ids
 
-    def test_sanitizer_matches_responses_call_id_when_id_differs(self, compressor):
-        msgs = [
-            {
-                "role": "assistant",
-                "content": "",
-                "tool_calls": [
-                    {
-                        "id": "fc_123",
-                        "call_id": "call_123",
-                        "response_item_id": "fc_123",
-                        "type": "function",
-                        "function": {"name": "search_files", "arguments": "{}"},
-                    }
-                ],
-            },
-            {"role": "tool", "tool_call_id": "call_123", "content": "result"},
-        ]
-
-        sanitized = compressor._sanitize_tool_pairs(msgs)
-
-        assert [m.get("tool_call_id") for m in sanitized if m.get("role") == "tool"] == [
-            "call_123"
-        ]
-
     def test_summary_role_avoids_consecutive_user_messages(self):
         """Summary role should alternate with the last head message to avoid consecutive same-role messages."""
         mock_client = MagicMock()
@@ -1143,34 +1119,6 @@ class TestTokenBudgetTailProtection:
         # At least one old tool result should have been pruned
         assert pruned >= 1
 
-    def test_prune_short_conv_protects_entire_tail(self, budget_compressor):
-        """Regression guard for PR #17025.
-
-        When ``len(messages) <= protect_tail_count`` and a token budget is
-        also set, every message must be protected. The previous code used
-        ``min(protect_tail_count, len(result) - 1)`` which capped the floor
-        one below the full length, leaving the oldest message eligible for
-        pruning.
-        """
-        c = budget_compressor
-        # 4 messages, protect_tail_count=4 -- nothing should be pruned.
-        # Oldest message is a large tool result; on the buggy path it falls
-        # outside the protected window and gets summarized.
-        messages = [
-            {"role": "tool", "content": "x" * 5000, "tool_call_id": "c0"},
-            {"role": "assistant", "content": "ack"},
-            {"role": "user", "content": "recent"},
-            {"role": "assistant", "content": "reply"},
-        ]
-        result, pruned = c._prune_old_tool_results(
-            messages,
-            protect_tail_count=4,
-            protect_tail_tokens=1_000_000,  # budget large enough to protect all
-        )
-        assert pruned == 0
-        # Tool result at index 0 must be preserved verbatim
-        assert result[0]["content"] == "x" * 5000
-
     def test_prune_without_token_budget_uses_message_count(self, budget_compressor):
         """Without protect_tail_tokens, falls back to message-count behavior."""
         c = budget_compressor

tests/agent/test_curator.py

@@ -363,19 +363,6 @@ def test_state_atomic_write_no_tmp_leftovers(curator_env):
         assert not p.name.startswith(".curator_state_"), f"tmp leftover: {p.name}"
 
 
-def test_state_preserves_last_report_path(curator_env):
-    c = curator_env["curator"]
-    c.save_state({
-        "last_run_at": "2026-04-30T12:00:00+00:00",
-        "last_run_summary": "ok",
-        "last_report_path": "/tmp/curator-report",
-        "paused": False,
-        "run_count": 1,
-    })
-    state = c.load_state()
-    assert state["last_report_path"] == "/tmp/curator-report"
-
-
 def test_curator_review_prompt_has_invariants():
     """Core invariants must be in the review prompt text."""
     from agent.curator import CURATOR_REVIEW_PROMPT

tests/agent/test_curator_reports.py

@@ -270,167 +270,3 @@ def test_state_transitions_captured_in_report(curator_env):
     assert "State transitions" in md
     assert "getting-old" in md
     assert "active → stale" in md
-
-
-# ---------------------------------------------------------------------------
-# Cron job skill reference rewriting (curator ↔ cron integration)
-# ---------------------------------------------------------------------------
-#
-# When the curator consolidates skill X into umbrella Y during a run, any
-# cron job that listed X in its ``skills`` field would fail to load X at
-# run time — the scheduler logs a warning and skips it, so the scheduled
-# job runs without the instructions it was scheduled to follow. These
-# tests verify that _write_run_report calls into cron.jobs to repair
-# those references and records what it did in both run.json and
-# cron_rewrites.json.
-
-
-@pytest.fixture
-def curator_env_with_cron(curator_env, monkeypatch):
-    """Extend curator_env with an initialized + repointed cron.jobs module."""
-    home = curator_env["home"]
-    (home / "cron").mkdir(exist_ok=True)
-    (home / "cron" / "output").mkdir(exist_ok=True)
-
-    import importlib
-    import cron.jobs as jobs_mod
-    importlib.reload(jobs_mod)
-    monkeypatch.setattr(jobs_mod, "HERMES_DIR", home)
-    monkeypatch.setattr(jobs_mod, "CRON_DIR", home / "cron")
-    monkeypatch.setattr(jobs_mod, "JOBS_FILE", home / "cron" / "jobs.json")
-    monkeypatch.setattr(jobs_mod, "OUTPUT_DIR", home / "cron" / "output")
-
-    return {**curator_env, "jobs": jobs_mod}
-
-
-def test_curator_rewrites_cron_skills_when_skill_consolidated(curator_env_with_cron):
-    """A skill consolidated into an umbrella should be rewritten in any
-    cron job's skills list; the rewrite should be visible in run.json
-    and cron_rewrites.json."""
-    curator = curator_env_with_cron["curator"]
-    jobs = curator_env_with_cron["jobs"]
-
-    # Create a cron job that depends on a soon-to-be-consolidated skill
-    job = jobs.create_job(
-        prompt="",
-        schedule="every 1h",
-        skills=["foo"],
-        name="foo-watcher",
-    )
-
-    # Simulate a curator pass that consolidated `foo` → `foo-umbrella`
-    before = [{"name": "foo", "state": "active", "pinned": False}]
-    after = [{"name": "foo-umbrella", "state": "active", "pinned": False}]
-
-    run_dir = curator._write_run_report(
-        started_at=datetime.now(timezone.utc),
-        elapsed_seconds=3.0,
-        auto_counts={"checked": 1, "marked_stale": 0, "archived": 0, "reactivated": 0},
-        auto_summary="no changes",
-        before_report=before,
-        before_names={"foo"},
-        after_report=after,
-        llm_meta=_make_llm_meta(
-            final="Consolidated foo into foo-umbrella.",
-            tool_calls=[
-                {
-                    "name": "skill_manage",
-                    "arguments": json.dumps({
-                        "action": "write_file",
-                        "name": "foo-umbrella",
-                        "file_path": "references/foo.md",
-                        "file_content": "from foo",
-                    }),
-                },
-            ],
-        ),
-    )
-
-    # Cron job is rewritten on disk
-    loaded = jobs.get_job(job["id"])
-    assert loaded["skills"] == ["foo-umbrella"]
-    assert loaded["skill"] == "foo-umbrella"
-
-    # Rewrite is recorded in run.json
-    payload = json.loads((run_dir / "run.json").read_text())
-    assert payload["cron_rewrites"]["jobs_updated"] == 1
-    assert payload["counts"]["cron_jobs_rewritten"] == 1
-    rewrites = payload["cron_rewrites"]["rewrites"]
-    assert len(rewrites) == 1
-    assert rewrites[0]["mapped"] == {"foo": "foo-umbrella"}
-
-    # Separate cron_rewrites.json is written for convenience
-    cron_file = run_dir / "cron_rewrites.json"
-    assert cron_file.exists()
-    detail = json.loads(cron_file.read_text())
-    assert detail["jobs_updated"] == 1
-
-    # Markdown surfaces the change
-    md = (run_dir / "REPORT.md").read_text()
-    assert "Cron job skill references rewritten" in md
-    assert "foo-watcher" in md
-    assert "foo-umbrella" in md
-
-
-def test_curator_drops_pruned_skill_from_cron_job(curator_env_with_cron):
-    """A pruned (no-umbrella) skill should be dropped from the cron
-    job's skill list entirely — there's no forwarding target."""
-    curator = curator_env_with_cron["curator"]
-    jobs = curator_env_with_cron["jobs"]
-
-    job = jobs.create_job(
-        prompt="",
-        schedule="every 1h",
-        skills=["keep", "stale-one"],
-    )
-
-    before = [{"name": "stale-one", "state": "active", "pinned": False}]
-    after: list = []  # stale-one was archived with no target
-
-    run_dir = curator._write_run_report(
-        started_at=datetime.now(timezone.utc),
-        elapsed_seconds=1.0,
-        auto_counts={"checked": 1, "marked_stale": 0, "archived": 1, "reactivated": 0},
-        auto_summary="1 archived",
-        before_report=before,
-        before_names={"stale-one"},
-        after_report=after,
-        llm_meta=_make_llm_meta(),  # no tool calls → classifier marks it pruned
-    )
-
-    loaded = jobs.get_job(job["id"])
-    assert loaded["skills"] == ["keep"]
-
-    payload = json.loads((run_dir / "run.json").read_text())
-    assert payload["cron_rewrites"]["jobs_updated"] == 1
-    rewrites = payload["cron_rewrites"]["rewrites"]
-    assert rewrites[0]["dropped"] == ["stale-one"]
-
-
-def test_curator_report_has_no_cron_section_when_nothing_changes(curator_env_with_cron):
-    """When the curator run doesn't touch any skills, cron jobs are
-    untouched and cron_rewrites.json is not even written."""
-    curator = curator_env_with_cron["curator"]
-    jobs = curator_env_with_cron["jobs"]
-
-    jobs.create_job(prompt="", schedule="every 1h", skills=["foo"])
-
-    run_dir = curator._write_run_report(
-        started_at=datetime.now(timezone.utc),
-        elapsed_seconds=1.0,
-        auto_counts={"checked": 0, "marked_stale": 0, "archived": 0, "reactivated": 0},
-        auto_summary="no changes",
-        before_report=[{"name": "foo", "state": "active", "pinned": False}],
-        before_names={"foo"},
-        after_report=[{"name": "foo", "state": "active", "pinned": False}],
-        llm_meta=_make_llm_meta(),
-    )
-
-    # No rewrites → no separate file, no section in md
-    assert not (run_dir / "cron_rewrites.json").exists()
-    md = (run_dir / "REPORT.md").read_text()
-    assert "Cron job skill references rewritten" not in md
-
-    payload = json.loads((run_dir / "run.json").read_text())
-    assert payload["cron_rewrites"]["jobs_updated"] == 0
-    assert payload["counts"]["cron_jobs_rewritten"] == 0

tests/agent/test_moonshot_schema.py

@@ -6,6 +6,11 @@ the JSON Schema ecosystem accepts:
 1. Properties without ``type`` — Moonshot requires ``type`` on every node.
 2. ``type`` at the parent of ``anyOf`` — Moonshot requires it only inside
    ``anyOf`` children.
+3. ``$ref`` with sibling keywords — Moonshot expands the ref first and then
+   rejects ``description``/``type`` siblings on the same node.
+   (Ported from anomalyco/opencode#24730.)
+4. Tuple-style ``items`` arrays — Moonshot requires a single item schema,
+   not positional ones. (Ported from anomalyco/opencode#24730.)
 
 These tests cover the repairs applied by ``agent/moonshot_schema.py``.
 """
@@ -115,15 +120,9 @@ class TestMissingTypeFilled:
 
 
 class TestAnyOfParentType:
-    """Rule 2: type must not appear at the anyOf parent level.
+    """Rule 2: type must not appear at the anyOf parent level."""
 
-    When an anyOf contains a null-type branch, Moonshot rejects it.
-    The sanitizer collapses the anyOf: single non-null branch is promoted,
-    multiple non-null branches have null removed from the list.
-    """
-
-    def test_anyof_null_branch_collapsed_to_single_type(self):
-        """anyOf [string, null] → plain string (anyOf removed)."""
+    def test_parent_type_stripped_when_anyof_present(self):
         params = {
             "type": "object",
             "properties": {
@@ -138,46 +137,179 @@ class TestAnyOfParentType:
         }
         out = sanitize_moonshot_tool_parameters(params)
         from_format = out["properties"]["from_format"]
-        # null branch removed, anyOf collapsed to the single non-null type
-        assert "anyOf" not in from_format
-        assert from_format["type"] == "string"
+        assert "type" not in from_format
+        assert "anyOf" in from_format
 
-    def test_anyof_multiple_non_null_preserved(self):
-        """anyOf [string, integer] (no null) → kept as-is with parent type stripped."""
+    def test_anyof_children_missing_type_get_filled(self):
         params = {
             "type": "object",
             "properties": {
-                "mode": {
+                "value": {
                     "anyOf": [
                         {"type": "string"},
-                        {"type": "integer"},
+                        {"description": "A typeless option"},
                     ],
                 },
             },
         }
         out = sanitize_moonshot_tool_parameters(params)
-        mode = out["properties"]["mode"]
-        assert "anyOf" in mode
-        assert "type" not in mode  # parent type stripped
+        children = out["properties"]["value"]["anyOf"]
+        assert children[0]["type"] == "string"
+        assert "type" in children[1]
 
-    def test_anyof_enum_with_null_collapsed(self):
-        """anyOf [{enum: [...], type: string}, {type: null}] → enum + type only."""
+
+class TestRefSiblingStripping:
+    """Rule 3: ``$ref`` nodes may not carry sibling keywords on Moonshot.
+
+    Ported from anomalyco/opencode#24730.  The real-world failure was MCP tools
+    whose generated schemas put a ``description`` on a ``$ref`` property so the
+    model would see the field's human-readable hint.  The reference stays — the
+    referenced definition still owns the description (on the target node itself)
+    and still serves the model's context.
+    """
+
+    def test_description_sibling_stripped_from_ref(self):
         params = {
             "type": "object",
             "properties": {
-                "db_type": {
+                "variantOptions": {
+                    "$ref": "#/$defs/VariantOptions",
+                    "description": "Required. The variant options for generation.",
+                },
+            },
+            "$defs": {
+                "VariantOptions": {
+                    "type": "object",
+                    "properties": {},
+                    "description": "Configuration options.",
+                },
+            },
+        }
+        out = sanitize_moonshot_tool_parameters(params)
+        # Sibling stripped.
+        assert out["properties"]["variantOptions"] == {"$ref": "#/$defs/VariantOptions"}
+        # The target definition's own description is preserved — we only strip
+        # siblings ON the $ref node, not on the thing it points at.
+        assert out["$defs"]["VariantOptions"]["description"] == "Configuration options."
+
+    def test_multiple_siblings_all_stripped(self):
+        params = {
+            "type": "object",
+            "properties": {
+                "p": {
+                    "$ref": "#/$defs/T",
+                    "type": "object",
+                    "description": "x",
+                    "default": {},
+                    "title": "P",
+                },
+            },
+            "$defs": {"T": {"type": "object"}},
+        }
+        out = sanitize_moonshot_tool_parameters(params)
+        assert out["properties"]["p"] == {"$ref": "#/$defs/T"}
+
+    def test_ref_without_siblings_unchanged(self):
+        params = {
+            "type": "object",
+            "properties": {"p": {"$ref": "#/$defs/T"}},
+            "$defs": {"T": {"type": "object"}},
+        }
+        out = sanitize_moonshot_tool_parameters(params)
+        assert out["properties"]["p"] == {"$ref": "#/$defs/T"}
+
+    def test_ref_inside_anyof_children(self):
+        params = {
+            "type": "object",
+            "properties": {
+                "v": {
                     "anyOf": [
-                        {"enum": ["mysql", "postgresql", ""]},
+                        {"$ref": "#/$defs/A", "description": "variant A"},
                         {"type": "null"},
                     ],
                 },
             },
+            "$defs": {"A": {"type": "object"}},
         }
         out = sanitize_moonshot_tool_parameters(params)
-        db_type = out["properties"]["db_type"]
-        assert "anyOf" not in db_type
-        assert db_type["type"] == "string"
-        assert db_type["enum"] == ["mysql", "postgresql"]  # "" stripped by enum cleanup
+        children = out["properties"]["v"]["anyOf"]
+        assert children[0] == {"$ref": "#/$defs/A"}
+        assert children[1] == {"type": "null"}
+
+
+class TestTupleItems:
+    """Rule 4: tuple-style ``items`` arrays collapse to a single schema.
+
+    Ported from anomalyco/opencode#24730.  Moonshot's schema engine requires
+    ``items`` to be ONE schema object applied to every array element; tuple-
+    style positional item schemas are rejected.  We collapse to the first
+    element's schema (which is the "closest" interpretation of positional →
+    single) and drop the rest.
+    """
+
+    def test_tuple_items_collapsed_to_first(self):
+        params = {
+            "type": "object",
+            "properties": {
+                "renderedSize": {
+                    "type": "array",
+                    "items": [{"type": "number"}, {"type": "number"}],
+                    "minItems": 2,
+                    "maxItems": 2,
+                },
+            },
+        }
+        out = sanitize_moonshot_tool_parameters(params)
+        assert out["properties"]["renderedSize"]["items"] == {"type": "number"}
+        # Sibling constraints are preserved — only the tuple shape is repaired.
+        assert out["properties"]["renderedSize"]["minItems"] == 2
+
+    def test_empty_tuple_items_becomes_empty_schema(self):
+        # Empty tuple collapses to ``{}``; the generic repair then fills a
+        # synthetic ``type`` because Moonshot requires ``type`` on every
+        # schema node.  Either ``{}`` or ``{"type": "string"}`` is a valid
+        # final shape for Moonshot — both accept any string element — but we
+        # always go through ``_fill_missing_type`` so the result is fully
+        # well-formed without needing the consumer to patch it later.
+        params = {
+            "type": "object",
+            "properties": {
+                "things": {"type": "array", "items": []},
+            },
+        }
+        out = sanitize_moonshot_tool_parameters(params)
+        items = out["properties"]["things"]["items"]
+        # Must be a dict and must carry a ``type`` (the whole point of Rule 1).
+        assert isinstance(items, dict)
+        assert items.get("type")
+
+    def test_tuple_items_first_element_is_repaired(self):
+        # The first element itself has a missing type — it should be filled.
+        params = {
+            "type": "object",
+            "properties": {
+                "pair": {
+                    "type": "array",
+                    "items": [{"description": "first"}, {"description": "second"}],
+                },
+            },
+        }
+        out = sanitize_moonshot_tool_parameters(params)
+        # Repaired to a single schema with a synthetic type.
+        assert out["properties"]["pair"]["items"] == {
+            "description": "first",
+            "type": "string",
+        }
+
+    def test_single_schema_items_unchanged(self):
+        params = {
+            "type": "object",
+            "properties": {
+                "tags": {"type": "array", "items": {"type": "string"}},
+            },
+        }
+        out = sanitize_moonshot_tool_parameters(params)
+        assert out["properties"]["tags"]["items"] == {"type": "string"}
 
 
 class TestTopLevelGuarantees:
@@ -253,7 +385,7 @@ class TestRealWorldMCPShape:
     """End-to-end: a realistic MCP-style schema that used to 400 on Moonshot."""
 
     def test_combined_rewrites(self):
-        # Shape: missing type on a property, anyOf with parent type + null, array
+        # Shape: missing type on a property, anyOf with parent type, array
         # items without type — all in one tool.
         params = {
             "type": "object",
@@ -275,125 +407,7 @@ class TestRealWorldMCPShape:
         }
         out = sanitize_moonshot_tool_parameters(params)
         assert out["properties"]["query"]["type"] == "string"
-        # anyOf with null collapsed to plain type
-        assert "anyOf" not in out["properties"]["filter"]
-        assert out["properties"]["filter"]["type"] == "string"
+        assert "type" not in out["properties"]["filter"]
+        assert out["properties"]["filter"]["anyOf"][0]["type"] == "string"
         assert out["properties"]["tags"]["items"]["type"] == "string"
         assert out["required"] == ["query"]
-
-
-class TestEnumNullStripping:
-    """Rule 3: Moonshot rejects null/empty-string inside enum arrays."""
-
-    def test_enum_null_value_stripped(self):
-        """enum containing Python None must have it removed for Moonshot."""
-        params = {
-            "type": "object",
-            "properties": {
-                "db_type": {
-                    "type": "string",
-                    "enum": ["mysql", "postgresql", None],
-                },
-            },
-        }
-        out = sanitize_moonshot_tool_parameters(params)
-        db_type = out["properties"]["db_type"]
-        assert None not in db_type["enum"]
-        assert "mysql" in db_type["enum"]
-        assert "postgresql" in db_type["enum"]
-
-    def test_enum_empty_string_stripped(self):
-        """enum containing empty string '' must have it removed for Moonshot."""
-        params = {
-            "type": "object",
-            "properties": {
-                "db_type": {
-                    "type": "string",
-                    "enum": ["mysql", "postgresql", ""],
-                },
-            },
-        }
-        out = sanitize_moonshot_tool_parameters(params)
-        db_type = out["properties"]["db_type"]
-        assert "" not in db_type["enum"]
-        assert db_type["enum"] == ["mysql", "postgresql"]
-
-    def test_enum_all_null_becomes_no_enum(self):
-        """enum that only had null/empty values is dropped entirely."""
-        params = {
-            "type": "object",
-            "properties": {
-                "val": {
-                    "type": "string",
-                    "enum": [None, ""],
-                },
-            },
-        }
-        out = sanitize_moonshot_tool_parameters(params)
-        assert "enum" not in out["properties"]["val"]
-
-    def test_dataslayer_db_type_after_mcp_normalize(self):
-        """Real-world: dataslayer db_type anyOf+enum after MCP normalization."""
-        # This is the exact shape after _normalize_mcp_input_schema runs:
-        # anyOf collapsed, but enum still has null + empty string
-        params = {
-            "type": "object",
-            "properties": {
-                "datasource": {"type": "string"},
-                "db_type": {
-                    "enum": ["mysql", "mariadb", "postgresql", "sqlserver", "oracle", "", None],
-                    "type": "string",
-                    "nullable": True,
-                    "default": None,
-                },
-            },
-            "required": ["datasource"],
-        }
-        out = sanitize_moonshot_tool_parameters(params)
-        db_type = out["properties"]["db_type"]
-        assert "nullable" not in db_type, "nullable keyword must be stripped"
-        assert None not in db_type["enum"]
-        assert "" not in db_type["enum"]
-        assert db_type["enum"] == ["mysql", "mariadb", "postgresql", "sqlserver", "oracle"]
-        assert db_type["type"] == "string"
-
-    def test_enum_on_object_type_not_stripped(self):
-        """enum on non-scalar types (object) should NOT be touched."""
-        params = {
-            "type": "object",
-            "properties": {
-                "config": {
-                    "type": "object",
-                    "properties": {},
-                    "enum": [{}, None],
-                },
-            },
-        }
-        out = sanitize_moonshot_tool_parameters(params)
-        # object-typed enum should pass through unchanged
-        assert "enum" in out["properties"]["config"]
-
-    def test_anyof_collapse_still_runs_nullable_and_enum_cleanup(self):
-        """After anyOf collapses to a single non-null branch, the merged
-        node must still have ``nullable`` stripped and null/empty-string
-        values removed from enum — not skipped by the early anyOf return.
-        """
-        params = {
-            "type": "object",
-            "properties": {
-                "db_type": {
-                    "anyOf": [
-                        {"enum": ["mysql", "postgresql", "", None]},
-                        {"type": "null"},
-                    ],
-                    "nullable": True,
-                },
-            },
-        }
-        out = sanitize_moonshot_tool_parameters(params)
-        db_type = out["properties"]["db_type"]
-        assert "anyOf" not in db_type
-        assert "nullable" not in db_type, "nullable must be stripped after anyOf collapse"
-        assert db_type["type"] == "string"
-        assert db_type["enum"] == ["mysql", "postgresql"], \
-            "null/empty enum values must be stripped after anyOf collapse"

tests/agent/test_skill_utils.py

@@ -1,58 +0,0 @@
-"""Tests for agent/skill_utils.py — extract_skill_conditions metadata handling."""
-
-from agent.skill_utils import extract_skill_conditions
-
-
-def test_metadata_as_dict_with_hermes():
-    """Normal case: metadata is a dict containing hermes keys."""
-    frontmatter = {
-        "metadata": {
-            "hermes": {
-                "fallback_for_toolsets": ["toolset_a"],
-                "requires_toolsets": ["toolset_b"],
-                "fallback_for_tools": ["tool_x"],
-                "requires_tools": ["tool_y"],
-            }
-        }
-    }
-    result = extract_skill_conditions(frontmatter)
-    assert result["fallback_for_toolsets"] == ["toolset_a"]
-    assert result["requires_toolsets"] == ["toolset_b"]
-    assert result["fallback_for_tools"] == ["tool_x"]
-    assert result["requires_tools"] == ["tool_y"]
-
-
-def test_metadata_as_string_does_not_crash():
-    """Bug case: metadata is a non-dict truthy value (e.g. a YAML string)."""
-    frontmatter = {"metadata": "some text"}
-    result = extract_skill_conditions(frontmatter)
-    assert result == {
-        "fallback_for_toolsets": [],
-        "requires_toolsets": [],
-        "fallback_for_tools": [],
-        "requires_tools": [],
-    }
-
-
-def test_metadata_as_none():
-    """metadata key is present but set to null/None."""
-    frontmatter = {"metadata": None}
-    result = extract_skill_conditions(frontmatter)
-    assert result == {
-        "fallback_for_toolsets": [],
-        "requires_toolsets": [],
-        "fallback_for_tools": [],
-        "requires_tools": [],
-    }
-
-
-def test_metadata_missing_entirely():
-    """metadata key is absent from frontmatter."""
-    frontmatter = {"name": "my-skill", "description": "Does stuff."}
-    result = extract_skill_conditions(frontmatter)
-    assert result == {
-        "fallback_for_toolsets": [],
-        "requires_toolsets": [],
-        "fallback_for_tools": [],
-        "requires_tools": [],
-    }

tests/agent/test_tool_guardrails.py

@@ -1,238 +0,0 @@
-"""Pure tool-call guardrail primitive tests."""
-
-import json
-
-from agent.tool_guardrails import (
-    ToolCallGuardrailConfig,
-    ToolCallGuardrailController,
-    ToolCallSignature,
-    canonical_tool_args,
-)
-
-
-def test_tool_call_signature_hashes_canonical_nested_unicode_args_without_exposing_raw_args():
-    args_a = {
-        "z": [{"β": "☤", "a": 1}],
-        "a": {"y": 2, "x": "secret-token-value"},
-    }
-    args_b = {
-        "a": {"x": "secret-token-value", "y": 2},
-        "z": [{"a": 1, "β": "☤"}],
-    }
-
-    assert canonical_tool_args(args_a) == canonical_tool_args(args_b)
-    sig_a = ToolCallSignature.from_call("web_search", args_a)
-    sig_b = ToolCallSignature.from_call("web_search", args_b)
-
-    assert sig_a == sig_b
-    assert len(sig_a.args_hash) == 64
-    metadata = sig_a.to_metadata()
-    assert metadata == {"tool_name": "web_search", "args_hash": sig_a.args_hash}
-    assert "secret-token-value" not in json.dumps(metadata)
-    assert "☤" not in json.dumps(metadata)
-
-
-def test_default_config_is_soft_warning_only_with_hard_stop_disabled():
-    cfg = ToolCallGuardrailConfig()
-
-    assert cfg.warnings_enabled is True
-    assert cfg.hard_stop_enabled is False
-    assert cfg.exact_failure_warn_after == 2
-    assert cfg.same_tool_failure_warn_after == 3
-    assert cfg.no_progress_warn_after == 2
-    assert cfg.exact_failure_block_after == 5
-    assert cfg.same_tool_failure_halt_after == 8
-    assert cfg.no_progress_block_after == 5
-
-
-def test_config_parses_nested_warn_and_hard_stop_thresholds():
-    cfg = ToolCallGuardrailConfig.from_mapping(
-        {
-            "warnings_enabled": False,
-            "hard_stop_enabled": True,
-            "warn_after": {
-                "exact_failure": 3,
-                "same_tool_failure": 4,
-                "idempotent_no_progress": 5,
-            },
-            "hard_stop_after": {
-                "exact_failure": 6,
-                "same_tool_failure": 7,
-                "idempotent_no_progress": 8,
-            },
-        }
-    )
-
-    assert cfg.warnings_enabled is False
-    assert cfg.hard_stop_enabled is True
-    assert cfg.exact_failure_warn_after == 3
-    assert cfg.same_tool_failure_warn_after == 4
-    assert cfg.no_progress_warn_after == 5
-    assert cfg.exact_failure_block_after == 6
-    assert cfg.same_tool_failure_halt_after == 7
-    assert cfg.no_progress_block_after == 8
-
-
-def test_default_repeated_identical_failed_call_warns_without_blocking():
-    controller = ToolCallGuardrailController()
-    args = {"query": "same"}
-
-    decisions = []
-    for _ in range(5):
-        assert controller.before_call("web_search", args).action == "allow"
-        decisions.append(
-            controller.after_call("web_search", args, '{"error":"boom"}', failed=True)
-        )
-
-    assert decisions[0].action == "allow"
-    assert [d.action for d in decisions[1:]] == ["warn", "warn", "warn", "warn"]
-    assert {d.code for d in decisions[1:]} == {"repeated_exact_failure_warning"}
-    assert controller.before_call("web_search", args).action == "allow"
-    assert controller.halt_decision is None
-
-
-def test_hard_stop_enabled_blocks_repeated_exact_failure_before_next_execution():
-    controller = ToolCallGuardrailController(
-        ToolCallGuardrailConfig(
-            hard_stop_enabled=True,
-            exact_failure_warn_after=2,
-            exact_failure_block_after=2,
-            same_tool_failure_halt_after=99,
-        )
-    )
-    args = {"query": "same"}
-
-    assert controller.before_call("web_search", args).action == "allow"
-    first = controller.after_call("web_search", args, '{"error":"boom"}', failed=True)
-    assert first.action == "allow"
-
-    assert controller.before_call("web_search", args).action == "allow"
-    second = controller.after_call("web_search", args, '{"error":"boom"}', failed=True)
-    assert second.action == "warn"
-    assert second.code == "repeated_exact_failure_warning"
-
-    blocked = controller.before_call("web_search", args)
-    assert blocked.action == "block"
-    assert blocked.code == "repeated_exact_failure_block"
-    assert blocked.count == 2
-
-
-def test_success_resets_exact_signature_failure_streak():
-    controller = ToolCallGuardrailController(
-        ToolCallGuardrailConfig(hard_stop_enabled=True, exact_failure_block_after=2, same_tool_failure_halt_after=99)
-    )
-    args = {"query": "same"}
-
-    controller.after_call("web_search", args, '{"error":"boom"}', failed=True)
-    controller.after_call("web_search", args, '{"ok":true}', failed=False)
-
-    assert controller.before_call("web_search", args).action == "allow"
-    controller.after_call("web_search", args, '{"error":"boom"}', failed=True)
-    assert controller.before_call("web_search", args).action == "allow"
-
-
-def test_same_tool_varying_args_warns_by_default_without_halting():
-    controller = ToolCallGuardrailController(
-        ToolCallGuardrailConfig(same_tool_failure_warn_after=2, same_tool_failure_halt_after=3)
-    )
-
-    first = controller.after_call("terminal", {"command": "cmd-1"}, '{"exit_code":1}', failed=True)
-    second = controller.after_call("terminal", {"command": "cmd-2"}, '{"exit_code":1}', failed=True)
-    third = controller.after_call("terminal", {"command": "cmd-3"}, '{"exit_code":1}', failed=True)
-    fourth = controller.after_call("terminal", {"command": "cmd-4"}, '{"exit_code":1}', failed=True)
-
-    assert first.action == "allow"
-    assert [second.action, third.action, fourth.action] == ["warn", "warn", "warn"]
-    assert {second.code, third.code, fourth.code} == {"same_tool_failure_warning"}
-    assert controller.halt_decision is None
-
-
-def test_hard_stop_enabled_halts_same_tool_varying_args_failure_streak():
-    controller = ToolCallGuardrailController(
-        ToolCallGuardrailConfig(
-            hard_stop_enabled=True,
-            exact_failure_block_after=99,
-            same_tool_failure_warn_after=2,
-            same_tool_failure_halt_after=3,
-        )
-    )
-
-    first = controller.after_call("terminal", {"command": "cmd-1"}, '{"exit_code":1}', failed=True)
-    assert first.action == "allow"
-    second = controller.after_call("terminal", {"command": "cmd-2"}, '{"exit_code":1}', failed=True)
-    assert second.action == "warn"
-    assert second.code == "same_tool_failure_warning"
-    third = controller.after_call("terminal", {"command": "cmd-3"}, '{"exit_code":1}', failed=True)
-    assert third.action == "halt"
-    assert third.code == "same_tool_failure_halt"
-    assert third.count == 3
-
-
-def test_idempotent_no_progress_repeated_result_warns_without_blocking_by_default():
-    controller = ToolCallGuardrailController(
-        ToolCallGuardrailConfig(no_progress_warn_after=2, no_progress_block_after=2)
-    )
-    args = {"path": "/tmp/same.txt"}
-    result = "same file contents"
-
-    for _ in range(4):
-        assert controller.before_call("read_file", args).action == "allow"
-        decision = controller.after_call("read_file", args, result, failed=False)
-
-    assert decision.action == "warn"
-    assert decision.code == "idempotent_no_progress_warning"
-    assert controller.before_call("read_file", args).action == "allow"
-    assert controller.halt_decision is None
-
-
-def test_hard_stop_enabled_blocks_idempotent_no_progress_future_repeat():
-    controller = ToolCallGuardrailController(
-        ToolCallGuardrailConfig(
-            hard_stop_enabled=True,
-            no_progress_warn_after=2,
-            no_progress_block_after=2,
-        )
-    )
-    args = {"path": "/tmp/same.txt"}
-    result = "same file contents"
-
-    assert controller.before_call("read_file", args).action == "allow"
-    assert controller.after_call("read_file", args, result, failed=False).action == "allow"
-    assert controller.before_call("read_file", args).action == "allow"
-    warn = controller.after_call("read_file", args, result, failed=False)
-    assert warn.action == "warn"
-    assert warn.code == "idempotent_no_progress_warning"
-
-    blocked = controller.before_call("read_file", args)
-    assert blocked.action == "block"
-    assert blocked.code == "idempotent_no_progress_block"
-
-
-def test_mutating_or_unknown_tools_are_not_blocked_for_repeated_identical_success_output_by_default():
-    controller = ToolCallGuardrailController(
-        ToolCallGuardrailConfig(no_progress_warn_after=2, no_progress_block_after=2)
-    )
-
-    for _ in range(3):
-        assert controller.before_call("write_file", {"path": "/tmp/x", "content": "x"}).action == "allow"
-        assert controller.after_call("write_file", {"path": "/tmp/x", "content": "x"}, "ok", failed=False).action == "allow"
-        assert controller.before_call("custom_tool", {"x": 1}).action == "allow"
-        assert controller.after_call("custom_tool", {"x": 1}, "ok", failed=False).action == "allow"
-
-
-def test_reset_for_turn_clears_bounded_guardrail_state():
-    controller = ToolCallGuardrailController(
-        ToolCallGuardrailConfig(hard_stop_enabled=True, exact_failure_block_after=2, no_progress_block_after=2)
-    )
-    controller.after_call("web_search", {"query": "same"}, '{"error":"boom"}', failed=True)
-    controller.after_call("web_search", {"query": "same"}, '{"error":"boom"}', failed=True)
-    controller.after_call("read_file", {"path": "/tmp/x"}, "same", failed=False)
-    controller.after_call("read_file", {"path": "/tmp/x"}, "same", failed=False)
-
-    assert controller.before_call("web_search", {"query": "same"}).action == "block"
-    assert controller.before_call("read_file", {"path": "/tmp/x"}).action == "block"
-
-    controller.reset_for_turn()
-
-    assert controller.before_call("web_search", {"query": "same"}).action == "allow"
-    assert controller.before_call("read_file", {"path": "/tmp/x"}).action == "allow"

tests/cli/test_manual_compress.py

@@ -21,21 +21,20 @@ def test_manual_compress_reports_noop_without_success_banner(capsys):
     shell.agent = MagicMock()
     shell.agent.compression_enabled = True
     shell.agent._cached_system_prompt = ""
-    shell.agent.tools = None
     shell.agent.session_id = shell.session_id  # no-op compression: no split
     shell.agent._compress_context.return_value = (list(history), "")
 
-    def _estimate(messages, **_kwargs):
+    def _estimate(messages):
         assert messages == history
         return 100
 
-    with patch("agent.model_metadata.estimate_request_tokens_rough", side_effect=_estimate):
+    with patch("agent.model_metadata.estimate_messages_tokens_rough", side_effect=_estimate):
         shell._manual_compress()
 
     output = capsys.readouterr().out
     assert "No changes from compression" in output
     assert "✅ Compressed" not in output
-    assert "Approx request size: ~100 tokens (unchanged)" in output
+    assert "Rough transcript estimate: ~100 tokens (unchanged)" in output
 
 
 def test_manual_compress_explains_when_token_estimate_rises(capsys):
@@ -50,23 +49,22 @@ def test_manual_compress_explains_when_token_estimate_rises(capsys):
     shell.agent = MagicMock()
     shell.agent.compression_enabled = True
     shell.agent._cached_system_prompt = ""
-    shell.agent.tools = None
     shell.agent.session_id = shell.session_id  # no-op: no split
     shell.agent._compress_context.return_value = (compressed, "")
 
-    def _estimate(messages, **_kwargs):
+    def _estimate(messages):
         if messages == history:
             return 100
         if messages == compressed:
             return 120
         raise AssertionError(f"unexpected transcript: {messages!r}")
 
-    with patch("agent.model_metadata.estimate_request_tokens_rough", side_effect=_estimate):
+    with patch("agent.model_metadata.estimate_messages_tokens_rough", side_effect=_estimate):
         shell._manual_compress()
 
     output = capsys.readouterr().out
     assert "✅ Compressed: 4 → 3 messages" in output
-    assert "Approx request size: ~100 → ~120 tokens" in output
+    assert "Rough transcript estimate: ~100 → ~120 tokens" in output
     assert "denser summaries" in output
 
 
@@ -91,7 +89,6 @@ def test_manual_compress_syncs_session_id_after_split():
     shell.agent = MagicMock()
     shell.agent.compression_enabled = True
     shell.agent._cached_system_prompt = ""
-    shell.agent.tools = None
     # Simulate _compress_context mutating agent.session_id as a side effect.
     def _fake_compress(*args, **kwargs):
         shell.agent.session_id = new_child_id
@@ -100,7 +97,7 @@ def test_manual_compress_syncs_session_id_after_split():
     shell.agent.session_id = old_id  # starts in sync
     shell._pending_title = "stale title"
 
-    with patch("agent.model_metadata.estimate_request_tokens_rough", return_value=100):
+    with patch("agent.model_metadata.estimate_messages_tokens_rough", return_value=100):
         shell._manual_compress()
 
     # CLI session_id must now point at the continuation child, not the parent.
@@ -121,12 +118,11 @@ def test_manual_compress_no_sync_when_session_id_unchanged():
     shell.agent = MagicMock()
     shell.agent.compression_enabled = True
     shell.agent._cached_system_prompt = ""
-    shell.agent.tools = None
     shell.agent.session_id = shell.session_id
     shell.agent._compress_context.return_value = (list(history), "")
     shell._pending_title = "keep me"
 
-    with patch("agent.model_metadata.estimate_request_tokens_rough", return_value=100):
+    with patch("agent.model_metadata.estimate_messages_tokens_rough", return_value=100):
         shell._manual_compress()
 
     # No split → pending title untouched.

tests/cron/test_rewrite_skill_refs.py

@@ -1,289 +0,0 @@
-"""Tests for cron.jobs.rewrite_skill_refs — the curator integration that
-keeps scheduled cron jobs pointing at the right skill names after a
-consolidation / pruning pass.
-
-Bug this fixes: when the curator consolidates skill X into umbrella Y,
-any cron job whose ``skills`` list contains X would silently fail to
-load X at run time (the scheduler logs a warning and skips it), so the
-job runs without the instructions it was scheduled to follow.
-"""
-
-from __future__ import annotations
-
-import sys
-from pathlib import Path
-
-import pytest
-
-# Ensure project root is importable
-sys.path.insert(0, str(Path(__file__).parent.parent.parent))
-
-
-@pytest.fixture
-def cron_env(tmp_path, monkeypatch):
-    """Isolated cron environment with temp HERMES_HOME."""
-    hermes_home = tmp_path / ".hermes"
-    hermes_home.mkdir()
-    (hermes_home / "cron").mkdir()
-    (hermes_home / "cron" / "output").mkdir()
-    monkeypatch.setenv("HERMES_HOME", str(hermes_home))
-
-    import cron.jobs as jobs_mod
-    monkeypatch.setattr(jobs_mod, "HERMES_DIR", hermes_home)
-    monkeypatch.setattr(jobs_mod, "CRON_DIR", hermes_home / "cron")
-    monkeypatch.setattr(jobs_mod, "JOBS_FILE", hermes_home / "cron" / "jobs.json")
-    monkeypatch.setattr(jobs_mod, "OUTPUT_DIR", hermes_home / "cron" / "output")
-
-    return hermes_home
-
-
-class TestRewriteSkillRefsNoop:
-    """No jobs, no rewrites, no map — every combination of empty inputs."""
-
-    def test_empty_map_and_no_jobs(self, cron_env):
-        from cron.jobs import rewrite_skill_refs
-
-        report = rewrite_skill_refs(consolidated={}, pruned=[])
-        assert report == {"rewrites": [], "jobs_updated": 0, "jobs_scanned": 0}
-
-    def test_jobs_exist_but_map_empty(self, cron_env):
-        from cron.jobs import create_job, rewrite_skill_refs
-
-        create_job(prompt="", schedule="every 1h", skills=["foo"])
-        report = rewrite_skill_refs(consolidated={}, pruned=[])
-        assert report["jobs_updated"] == 0
-        # Early return: we don't even scan when there's nothing to apply.
-        assert report["jobs_scanned"] == 0
-
-    def test_jobs_exist_but_no_match(self, cron_env):
-        from cron.jobs import create_job, get_job, rewrite_skill_refs
-
-        job = create_job(prompt="", schedule="every 1h", skills=["foo"])
-        report = rewrite_skill_refs(
-            consolidated={"unrelated": "umbrella"},
-            pruned=["other"],
-        )
-        assert report["jobs_updated"] == 0
-        assert report["jobs_scanned"] == 1
-        # Job untouched
-        loaded = get_job(job["id"])
-        assert loaded["skills"] == ["foo"]
-
-
-class TestRewriteSkillRefsConsolidation:
-    """Consolidated skills should be replaced with their umbrella target."""
-
-    def test_single_skill_replaced(self, cron_env):
-        from cron.jobs import create_job, get_job, rewrite_skill_refs
-
-        job = create_job(prompt="", schedule="every 1h", skills=["legacy-skill"])
-        report = rewrite_skill_refs(
-            consolidated={"legacy-skill": "umbrella-skill"},
-            pruned=[],
-        )
-
-        assert report["jobs_updated"] == 1
-        loaded = get_job(job["id"])
-        assert loaded["skills"] == ["umbrella-skill"]
-        # Legacy ``skill`` field realigned
-        assert loaded["skill"] == "umbrella-skill"
-
-    def test_multiple_skills_one_consolidated(self, cron_env):
-        from cron.jobs import create_job, get_job, rewrite_skill_refs
-
-        job = create_job(
-            prompt="",
-            schedule="every 1h",
-            skills=["keep-a", "legacy", "keep-b"],
-        )
-        rewrite_skill_refs(consolidated={"legacy": "umbrella"}, pruned=[])
-
-        loaded = get_job(job["id"])
-        # Ordering preserved, legacy replaced in-place
-        assert loaded["skills"] == ["keep-a", "umbrella", "keep-b"]
-
-    def test_umbrella_already_in_list_dedupes(self, cron_env):
-        from cron.jobs import create_job, get_job, rewrite_skill_refs
-
-        # Job already loads the umbrella AND the legacy sub-skill
-        job = create_job(
-            prompt="",
-            schedule="every 1h",
-            skills=["umbrella", "legacy"],
-        )
-        rewrite_skill_refs(consolidated={"legacy": "umbrella"}, pruned=[])
-
-        loaded = get_job(job["id"])
-        # No duplicate — the umbrella stays exactly once
-        assert loaded["skills"] == ["umbrella"]
-
-    def test_rewrite_report_records_mapping(self, cron_env):
-        from cron.jobs import create_job, rewrite_skill_refs
-
-        job = create_job(
-            prompt="",
-            schedule="every 1h",
-            skills=["a", "b"],
-            name="my-job",
-        )
-        report = rewrite_skill_refs(
-            consolidated={"a": "umbrella-a", "b": "umbrella-b"},
-            pruned=[],
-        )
-
-        assert len(report["rewrites"]) == 1
-        entry = report["rewrites"][0]
-        assert entry["job_id"] == job["id"]
-        assert entry["job_name"] == "my-job"
-        assert entry["before"] == ["a", "b"]
-        assert entry["after"] == ["umbrella-a", "umbrella-b"]
-        assert entry["mapped"] == {"a": "umbrella-a", "b": "umbrella-b"}
-        assert entry["dropped"] == []
-
-
-class TestRewriteSkillRefsPruning:
-    """Pruned skills should be dropped outright (no forwarding target)."""
-
-    def test_pruned_skill_dropped(self, cron_env):
-        from cron.jobs import create_job, get_job, rewrite_skill_refs
-
-        job = create_job(
-            prompt="",
-            schedule="every 1h",
-            skills=["keep", "stale"],
-        )
-        report = rewrite_skill_refs(consolidated={}, pruned=["stale"])
-
-        assert report["jobs_updated"] == 1
-        loaded = get_job(job["id"])
-        assert loaded["skills"] == ["keep"]
-        assert loaded["skill"] == "keep"
-
-    def test_all_skills_pruned_leaves_empty_list(self, cron_env):
-        from cron.jobs import create_job, get_job, rewrite_skill_refs
-
-        job = create_job(prompt="", schedule="every 1h", skills=["gone"])
-        rewrite_skill_refs(consolidated={}, pruned=["gone"])
-
-        loaded = get_job(job["id"])
-        assert loaded["skills"] == []
-        assert loaded["skill"] is None
-
-    def test_pruned_report_records_drops(self, cron_env):
-        from cron.jobs import create_job, rewrite_skill_refs
-
-        create_job(prompt="", schedule="every 1h", skills=["keep", "stale"])
-        report = rewrite_skill_refs(consolidated={}, pruned=["stale"])
-
-        entry = report["rewrites"][0]
-        assert entry["dropped"] == ["stale"]
-        assert entry["mapped"] == {}
-
-
-class TestRewriteSkillRefsMixed:
-    """Consolidation + pruning in the same pass."""
-
-    def test_mixed_consolidation_and_pruning(self, cron_env):
-        from cron.jobs import create_job, get_job, rewrite_skill_refs
-
-        job = create_job(
-            prompt="",
-            schedule="every 1h",
-            skills=["keep", "legacy", "stale"],
-        )
-        rewrite_skill_refs(
-            consolidated={"legacy": "umbrella"},
-            pruned=["stale"],
-        )
-
-        loaded = get_job(job["id"])
-        assert loaded["skills"] == ["keep", "umbrella"]
-
-    def test_skill_in_both_maps_wins_as_consolidated(self, cron_env):
-        """Defensive: if a skill appears in both lists (shouldn't happen
-        in practice), prefer consolidation — it has a forwarding target,
-        which is the more useful outcome."""
-        from cron.jobs import create_job, get_job, rewrite_skill_refs
-
-        job = create_job(prompt="", schedule="every 1h", skills=["ambiguous"])
-        rewrite_skill_refs(
-            consolidated={"ambiguous": "umbrella"},
-            pruned=["ambiguous"],
-        )
-
-        loaded = get_job(job["id"])
-        assert loaded["skills"] == ["umbrella"]
-
-
-class TestRewriteSkillRefsMultipleJobs:
-    """Multiple jobs, some affected, some not."""
-
-    def test_only_affected_jobs_reported(self, cron_env):
-        from cron.jobs import create_job, get_job, rewrite_skill_refs
-
-        j1 = create_job(prompt="", schedule="every 1h", skills=["legacy"])
-        j2 = create_job(prompt="", schedule="every 1h", skills=["untouched"])
-        j3 = create_job(prompt="", schedule="every 1h", skills=[])
-
-        report = rewrite_skill_refs(
-            consolidated={"legacy": "umbrella"},
-            pruned=[],
-        )
-
-        assert report["jobs_updated"] == 1
-        assert report["jobs_scanned"] == 3
-        assert len(report["rewrites"]) == 1
-        assert report["rewrites"][0]["job_id"] == j1["id"]
-
-        # Untouched jobs stay put
-        assert get_job(j2["id"])["skills"] == ["untouched"]
-        assert get_job(j3["id"])["skills"] == []
-
-    def test_legacy_skill_field_also_rewritten(self, cron_env):
-        """Old jobs may have the legacy single-skill ``skill`` field
-        set instead of ``skills``. Both paths should be rewritten."""
-        from cron.jobs import create_job, get_job, rewrite_skill_refs
-
-        # Create via the legacy ``skill`` argument
-        job = create_job(
-            prompt="",
-            schedule="every 1h",
-            skill="legacy",
-        )
-        rewrite_skill_refs(consolidated={"legacy": "umbrella"}, pruned=[])
-
-        loaded = get_job(job["id"])
-        assert loaded["skills"] == ["umbrella"]
-        assert loaded["skill"] == "umbrella"
-
-
-class TestRewriteSkillRefsPersistence:
-    """Rewrites persist to disk and survive a reload."""
-
-    def test_changes_persist_across_reload(self, cron_env):
-        import json
-        from cron.jobs import create_job, rewrite_skill_refs, JOBS_FILE
-
-        create_job(prompt="", schedule="every 1h", skills=["legacy"])
-        rewrite_skill_refs(consolidated={"legacy": "umbrella"}, pruned=[])
-
-        # Read raw file contents
-        data = json.loads(JOBS_FILE.read_text())
-        assert data["jobs"][0]["skills"] == ["umbrella"]
-        assert data["jobs"][0]["skill"] == "umbrella"
-
-    def test_noop_does_not_rewrite_file(self, cron_env):
-        from cron.jobs import create_job, rewrite_skill_refs, JOBS_FILE
-
-        create_job(prompt="", schedule="every 1h", skills=["keep"])
-        mtime_before = JOBS_FILE.stat().st_mtime_ns
-
-        # Nothing in the map matches
-        report = rewrite_skill_refs(
-            consolidated={"unrelated": "umbrella"},
-            pruned=["other"],
-        )
-
-        assert report["jobs_updated"] == 0
-        # File untouched — no pointless disk write
-        assert JOBS_FILE.stat().st_mtime_ns == mtime_before

tests/gateway/feishu_helpers.py

@@ -1,65 +0,0 @@
-"""Shared fixtures for Feishu adapter tests (admission, group policy, dispatch)."""
-
-from __future__ import annotations
-
-import threading
-from types import SimpleNamespace
-from typing import Any, Optional
-
-
-def make_sender(sender_type: str = "user", open_id: str = "ou_human",
-                user_id: Optional[str] = None, union_id: Optional[str] = None) -> Any:
-    return SimpleNamespace(
-        sender_type=sender_type,
-        sender_id=SimpleNamespace(open_id=open_id, user_id=user_id, union_id=union_id),
-    )
-
-
-def make_message(message_id: str = "om_xxx", chat_type: str = "p2p",
-                 chat_id: str = "oc_1", mentions: Optional[list] = None) -> Any:
-    return SimpleNamespace(
-        message_id=message_id,
-        chat_type=chat_type,
-        chat_id=chat_id,
-        mentions=mentions,
-        content="",
-        message_type="text",
-    )
-
-
-def make_adapter_skeleton(
-    *,
-    bot_open_id: str = "ou_me",
-    bot_user_id: str = "",
-    allow_bots: str = "none",
-    require_mention: bool = True,
-    group_policy: str = "allowlist",
-) -> Any:
-    from gateway.platforms.feishu import FeishuAdapter
-
-    adapter = object.__new__(FeishuAdapter)
-    adapter._bot_open_id = bot_open_id
-    adapter._bot_user_id = bot_user_id
-    adapter._bot_name = ""
-    adapter._app_id = ""
-    adapter._admins = set()
-    adapter._group_rules = {}
-    adapter._group_policy = group_policy
-    adapter._default_group_policy = group_policy
-    adapter._allowed_group_users = frozenset()
-    adapter._allow_bots = allow_bots
-    adapter._require_mention = require_mention
-    return adapter
-
-
-def install_dedup_state(adapter: Any, seen: Optional[dict] = None) -> None:
-    adapter._seen_message_ids = dict(seen) if seen else {}
-    adapter._seen_message_order = list((seen or {}).keys())
-    adapter._dedup_cache_size = 100
-    adapter._dedup_lock = threading.Lock()
-    adapter._dedup_state_path = None
-    adapter._persist_seen_message_ids = lambda: None
-
-
-def stub_mention(adapter: Any, mentions_self: bool) -> None:
-    adapter._mentions_self = lambda _message: mentions_self

tests/gateway/test_api_server.py

@@ -332,36 +332,6 @@ def auth_adapter():
     return _make_adapter(api_key="sk-secret")
 
 
-# ---------------------------------------------------------------------------
-# Adapter internals
-# ---------------------------------------------------------------------------
-
-
-class TestAgentExecution:
-    @pytest.mark.asyncio
-    async def test_run_agent_uses_session_id_as_task_id(self, adapter):
-        mock_agent = MagicMock()
-        mock_agent.run_conversation.return_value = {"final_response": "ok"}
-        mock_agent.session_prompt_tokens = 1
-        mock_agent.session_completion_tokens = 2
-        mock_agent.session_total_tokens = 3
-
-        with patch.object(adapter, "_create_agent", return_value=mock_agent):
-            result, usage = await adapter._run_agent(
-                user_message="hello",
-                conversation_history=[],
-                session_id="session-123",
-            )
-
-        assert result == {"final_response": "ok"}
-        assert usage == {"input_tokens": 1, "output_tokens": 2, "total_tokens": 3}
-        mock_agent.run_conversation.assert_called_once_with(
-            user_message="hello",
-            conversation_history=[],
-            task_id="session-123",
-        )
-
-
 # ---------------------------------------------------------------------------
 # /health endpoint
 # ---------------------------------------------------------------------------

tests/gateway/test_api_server_runs.py

@@ -253,7 +253,10 @@ class TestRunStatus:
                     await asyncio.sleep(0.05)
 
                 mock_agent.run_conversation.assert_called_once()
-                assert mock_agent.run_conversation.call_args.kwargs["task_id"] == "space-session"
+                # task_id stays "default" so the Runs API shares one sandbox
+                # container with CLI/gateway; session_id is surfaced in status
+                # for external UIs to correlate runs with their own session IDs.
+                assert mock_agent.run_conversation.call_args.kwargs["task_id"] == "default"
                 assert status["session_id"] == "space-session"
 
     @pytest.mark.asyncio

tests/gateway/test_approve_deny_commands.py

@@ -173,23 +173,6 @@ class TestBlockingGatewayApproval:
         assert e1.event.is_set()
         assert e2.event.is_set()
 
-    def test_clear_session_denies_and_signals_all_entries(self):
-        """clear_session must wake blocked entries during boundary cleanup."""
-        from tools.approval import clear_session, _ApprovalEntry, _gateway_queues
-
-        session_key = "test-boundary-cleanup"
-        e1 = _ApprovalEntry({"command": "cmd1"})
-        e2 = _ApprovalEntry({"command": "cmd2"})
-        _gateway_queues[session_key] = [e1, e2]
-
-        clear_session(session_key)
-
-        assert e1.event.is_set()
-        assert e2.event.is_set()
-        assert e1.result == "deny"
-        assert e2.result == "deny"
-        assert session_key not in _gateway_queues
-
 
 # ------------------------------------------------------------------
 # /approve command

tests/gateway/test_compress_command.py

@@ -64,13 +64,11 @@ async def test_compress_command_reports_noop_without_success_banner():
     agent_instance = MagicMock()
     agent_instance.shutdown_memory_provider = MagicMock()
     agent_instance.close = MagicMock()
-    agent_instance._cached_system_prompt = ""
-    agent_instance.tools = None
     agent_instance.context_compressor.has_content_to_compress.return_value = True
     agent_instance.session_id = "sess-1"
     agent_instance._compress_context.return_value = (list(history), "")
 
-    def _estimate(messages, **_kwargs):
+    def _estimate(messages):
         assert messages == history
         return 100
 
@@ -78,13 +76,13 @@ async def test_compress_command_reports_noop_without_success_banner():
         patch("gateway.run._resolve_runtime_agent_kwargs", return_value={"api_key": "test-key"}),
         patch("gateway.run._resolve_gateway_model", return_value="test-model"),
         patch("run_agent.AIAgent", return_value=agent_instance),
-        patch("agent.model_metadata.estimate_request_tokens_rough", side_effect=_estimate),
+        patch("agent.model_metadata.estimate_messages_tokens_rough", side_effect=_estimate),
     ):
         result = await runner._handle_compress_command(_make_event())
 
     assert "No changes from compression" in result
     assert "Compressed:" not in result
-    assert "Approx request size: ~100 tokens (unchanged)" in result
+    assert "Rough transcript estimate: ~100 tokens (unchanged)" in result
     agent_instance.shutdown_memory_provider.assert_called_once()
     agent_instance.close.assert_called_once()
 
@@ -101,13 +99,11 @@ async def test_compress_command_explains_when_token_estimate_rises():
     agent_instance = MagicMock()
     agent_instance.shutdown_memory_provider = MagicMock()
     agent_instance.close = MagicMock()
-    agent_instance._cached_system_prompt = ""
-    agent_instance.tools = None
     agent_instance.context_compressor.has_content_to_compress.return_value = True
     agent_instance.session_id = "sess-1"
     agent_instance._compress_context.return_value = (compressed, "")
 
-    def _estimate(messages, **_kwargs):
+    def _estimate(messages):
         if messages == history:
             return 100
         if messages == compressed:
@@ -118,12 +114,12 @@ async def test_compress_command_explains_when_token_estimate_rises():
         patch("gateway.run._resolve_runtime_agent_kwargs", return_value={"api_key": "test-key"}),
         patch("gateway.run._resolve_gateway_model", return_value="test-model"),
         patch("run_agent.AIAgent", return_value=agent_instance),
-        patch("agent.model_metadata.estimate_request_tokens_rough", side_effect=_estimate),
+        patch("agent.model_metadata.estimate_messages_tokens_rough", side_effect=_estimate),
     ):
         result = await runner._handle_compress_command(_make_event())
 
     assert "Compressed: 4 → 3 messages" in result
-    assert "Approx request size: ~100 → ~120 tokens" in result
+    assert "Rough transcript estimate: ~100 → ~120 tokens" in result
     assert "denser summaries" in result
     agent_instance.shutdown_memory_provider.assert_called_once()
     agent_instance.close.assert_called_once()
@@ -147,8 +143,6 @@ async def test_compress_command_appends_warning_when_summary_generation_fails():
     agent_instance = MagicMock()
     agent_instance.shutdown_memory_provider = MagicMock()
     agent_instance.close = MagicMock()
-    agent_instance._cached_system_prompt = ""
-    agent_instance.tools = None
     agent_instance.context_compressor.has_content_to_compress.return_value = True
     # Simulate summary-generation failure: fallback flag set, dropped count
     # populated, error string captured.
@@ -160,7 +154,7 @@ async def test_compress_command_appends_warning_when_summary_generation_fails():
     agent_instance.session_id = "sess-1"
     agent_instance._compress_context.return_value = (compressed, "")
 
-    def _estimate(messages, **_kwargs):
+    def _estimate(messages):
         if messages == history:
             return 100
         if messages == compressed:
@@ -171,7 +165,7 @@ async def test_compress_command_appends_warning_when_summary_generation_fails():
         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_request_tokens_rough", side_effect=_estimate),
+        patch("agent.model_metadata.estimate_messages_tokens_rough", side_effect=_estimate),
     ):
         result = await runner._handle_compress_command(_make_event())
 
@@ -206,8 +200,6 @@ async def test_compress_command_surfaces_aux_model_failure_even_when_recovered()
     agent_instance = MagicMock()
     agent_instance.shutdown_memory_provider = MagicMock()
     agent_instance.close = MagicMock()
-    agent_instance._cached_system_prompt = ""
-    agent_instance.tools = None
     agent_instance.context_compressor.has_content_to_compress.return_value = True
     # Fallback placeholder was NOT used — recovery succeeded.
     agent_instance.context_compressor._last_summary_fallback_used = False
@@ -223,7 +215,7 @@ async def test_compress_command_surfaces_aux_model_failure_even_when_recovered()
     agent_instance.session_id = "sess-1"
     agent_instance._compress_context.return_value = (compressed, "")
 
-    def _estimate(messages, **_kwargs):
+    def _estimate(messages):
         if messages == history:
             return 100
         if messages == compressed:
@@ -234,7 +226,7 @@ async def test_compress_command_surfaces_aux_model_failure_even_when_recovered()
         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_request_tokens_rough", side_effect=_estimate),
+        patch("agent.model_metadata.estimate_messages_tokens_rough", side_effect=_estimate),
     ):
         result = await runner._handle_compress_command(_make_event())
 

tests/gateway/test_config.py

@@ -9,7 +9,6 @@ from gateway.config import (
     Platform,
     PlatformConfig,
     SessionResetPolicy,
-    StreamingConfig,
     _apply_env_overrides,
     load_gateway_config,
 )
@@ -150,24 +149,6 @@ class TestSessionResetPolicy:
         assert restored.notify is False
 
 
-class TestStreamingConfig:
-    def test_from_dict_coerces_quoted_false_enabled(self):
-        restored = StreamingConfig.from_dict({"enabled": "false"})
-        assert restored.enabled is False
-
-    def test_from_dict_malformed_numeric_values_fall_back_to_defaults(self):
-        restored = StreamingConfig.from_dict(
-            {
-                "edit_interval": "oops",
-                "buffer_threshold": "oops",
-                "fresh_final_after_seconds": "oops",
-            }
-        )
-        assert restored.edit_interval == 1.0
-        assert restored.buffer_threshold == 40
-        assert restored.fresh_final_after_seconds == 60.0
-
-
 class TestGatewayConfigRoundtrip:
     def test_full_roundtrip(self):
         config = GatewayConfig(
@@ -379,38 +360,6 @@ class TestLoadGatewayConfig:
             "C01ABC": "Code review mode",
         }
 
-    def test_bridges_feishu_allow_bots_from_config_yaml_to_env(self, tmp_path, monkeypatch):
-        hermes_home = tmp_path / ".hermes"
-        hermes_home.mkdir()
-        config_path = hermes_home / "config.yaml"
-        config_path.write_text(
-            "feishu:\n  allow_bots: mentions\n",
-            encoding="utf-8",
-        )
-
-        monkeypatch.setenv("HERMES_HOME", str(hermes_home))
-        monkeypatch.delenv("FEISHU_ALLOW_BOTS", raising=False)
-
-        load_gateway_config()
-
-        assert os.environ.get("FEISHU_ALLOW_BOTS") == "mentions"
-
-    def test_feishu_allow_bots_env_takes_precedence_over_config_yaml(self, tmp_path, monkeypatch):
-        hermes_home = tmp_path / ".hermes"
-        hermes_home.mkdir()
-        config_path = hermes_home / "config.yaml"
-        config_path.write_text(
-            "feishu:\n  allow_bots: all\n",
-            encoding="utf-8",
-        )
-
-        monkeypatch.setenv("HERMES_HOME", str(hermes_home))
-        monkeypatch.setenv("FEISHU_ALLOW_BOTS", "none")
-
-        load_gateway_config()
-
-        assert os.environ.get("FEISHU_ALLOW_BOTS") == "none"
-
     def test_invalid_quick_commands_in_config_yaml_are_ignored(self, tmp_path, monkeypatch):
         hermes_home = tmp_path / ".hermes"
         hermes_home.mkdir()
@@ -506,15 +455,6 @@ class TestHomeChannelEnvOverrides:
                 {"SLACK_HOME_CHANNEL": "C123", "SLACK_HOME_CHANNEL_NAME": "Ops"},
                 ("C123", "Ops"),
             ),
-            (
-                Platform.WHATSAPP,
-                PlatformConfig(enabled=True),
-                {
-                    "WHATSAPP_HOME_CHANNEL": "1234567890@lid",
-                    "WHATSAPP_HOME_CHANNEL_NAME": "Owner DM",
-                },
-                ("1234567890@lid", "Owner DM"),
-            ),
             (
                 Platform.SIGNAL,
                 PlatformConfig(

tests/gateway/test_ephemeral_reply.py

@@ -1,336 +0,0 @@
-"""Tests for EphemeralReply — system-notice auto-delete in gateway adapters.
-
-Slash-command handlers in ``gateway/run.py`` can return an
-``EphemeralReply`` wrapper to request auto-deletion of the reply message
-after a TTL.  The base adapter unwraps the sentinel before sending and
-schedules a detached delete task when the platform supports
-``delete_message``.
-
-Covered:
-
-1. ``_unwrap_ephemeral`` returns text + ttl for EphemeralReply, and
-   passes plain strings through unchanged.
-2. TTL is zeroed on platforms that don't override ``delete_message``
-   (silent degrade — message stays in place).
-3. TTL is honored on platforms that DO override ``delete_message``.
-4. ``_schedule_ephemeral_delete`` invokes ``delete_message`` after the
-   configured delay with the correct chat_id / message_id.
-5. ``_process_message_background`` sends the unwrapped text (not the
-   sentinel object) and schedules deletion when appropriate.
-6. The two busy-session bypass paths also unwrap + schedule.
-"""
-
-import asyncio
-from unittest.mock import AsyncMock, patch
-
-import pytest
-
-from gateway.config import Platform, PlatformConfig
-from gateway.platforms.base import (
-    BasePlatformAdapter,
-    EphemeralReply,
-    MessageEvent,
-    MessageType,
-    SendResult,
-)
-from gateway.session import SessionSource
-
-
-class _NoDeleteAdapter(BasePlatformAdapter):
-    """Adapter that does NOT override delete_message (silent degrade)."""
-
-    async def connect(self):
-        pass
-
-    async def disconnect(self):
-        pass
-
-    async def send(self, chat_id, content="", **kwargs):
-        return SendResult(success=True, message_id="m-1")
-
-    async def get_chat_info(self, chat_id):
-        return {}
-
-
-class _DeleteCapableAdapter(BasePlatformAdapter):
-    """Adapter that overrides delete_message (TTL honored)."""
-
-    def __init__(self, *a, **kw):
-        super().__init__(*a, **kw)
-        self.deleted: list[tuple[str, str]] = []
-
-    async def connect(self):
-        pass
-
-    async def disconnect(self):
-        pass
-
-    async def send(self, chat_id, content="", **kwargs):
-        return SendResult(success=True, message_id="m-2")
-
-    async def get_chat_info(self, chat_id):
-        return {}
-
-    async def delete_message(self, chat_id: str, message_id: str) -> bool:
-        self.deleted.append((chat_id, message_id))
-        return True
-
-
-def _no_delete_adapter():
-    return _NoDeleteAdapter(
-        PlatformConfig(enabled=True, token="t"), Platform.TELEGRAM
-    )
-
-
-def _delete_adapter():
-    return _DeleteCapableAdapter(
-        PlatformConfig(enabled=True, token="t"), Platform.TELEGRAM
-    )
-
-
-def _make_event(text="/stop", chat_id="42"):
-    return MessageEvent(
-        text=text,
-        message_id="msg-1",
-        source=SessionSource(
-            platform=Platform.TELEGRAM,
-            chat_id=chat_id,
-            user_id="u-1",
-        ),
-        message_type=MessageType.TEXT,
-    )
-
-
-# ---------------------------------------------------------------------------
-# _unwrap_ephemeral
-# ---------------------------------------------------------------------------
-
-
-def test_unwrap_plain_string_is_passthrough():
-    adapter = _delete_adapter()
-    text, ttl = adapter._unwrap_ephemeral("hello")
-    assert text == "hello"
-    assert ttl == 0
-
-
-def test_unwrap_none_is_passthrough():
-    adapter = _delete_adapter()
-    text, ttl = adapter._unwrap_ephemeral(None)
-    assert text is None
-    assert ttl == 0
-
-
-def test_unwrap_ephemeral_explicit_ttl_on_capable_adapter():
-    adapter = _delete_adapter()
-    text, ttl = adapter._unwrap_ephemeral(EphemeralReply("bye", ttl_seconds=60))
-    assert text == "bye"
-    assert ttl == 60
-
-
-def test_unwrap_ephemeral_zeros_ttl_on_incapable_adapter():
-    """Platforms without delete_message should silently degrade to normal send."""
-    adapter = _no_delete_adapter()
-    text, ttl = adapter._unwrap_ephemeral(EphemeralReply("bye", ttl_seconds=60))
-    assert text == "bye"
-    assert ttl == 0  # forced to 0 — message will stay in place
-
-
-def test_unwrap_ephemeral_default_ttl_from_config():
-    adapter = _delete_adapter()
-    with patch.object(adapter, "_get_ephemeral_system_ttl_default", return_value=120):
-        text, ttl = adapter._unwrap_ephemeral(EphemeralReply("bye"))
-    assert text == "bye"
-    assert ttl == 120
-
-
-def test_unwrap_ephemeral_default_ttl_zero_disables():
-    """Config default of 0 (the shipped default) means the feature is off."""
-    adapter = _delete_adapter()
-    with patch.object(adapter, "_get_ephemeral_system_ttl_default", return_value=0):
-        text, ttl = adapter._unwrap_ephemeral(EphemeralReply("bye"))
-    assert text == "bye"
-    assert ttl == 0
-
-
-def test_unwrap_ephemeral_handles_unreadable_config():
-    adapter = _delete_adapter()
-    with patch.object(
-        adapter,
-        "_get_ephemeral_system_ttl_default",
-        side_effect=RuntimeError("boom"),
-    ):
-        text, ttl = adapter._unwrap_ephemeral(EphemeralReply("bye"))
-    # Fall back to 0 rather than crashing the handler pipeline.
-    assert text == "bye"
-    assert ttl == 0
-
-
-# ---------------------------------------------------------------------------
-# _schedule_ephemeral_delete
-# ---------------------------------------------------------------------------
-
-
-@pytest.mark.asyncio
-async def test_schedule_ephemeral_delete_calls_delete_after_ttl():
-    adapter = _delete_adapter()
-    # Use a very short TTL to keep the test fast — the implementation
-    # floors sleeps at 1s via ``max(1, int(ttl_seconds))``.  Patch asyncio.sleep
-    # inside the module under test; the test body uses the real one for
-    # scheduler pumping.
-    import gateway.platforms.base as base_module
-
-    sleeps: list[float] = []
-    _real_sleep = base_module.asyncio.sleep
-
-    async def _fake_sleep(duration):
-        sleeps.append(duration)
-        # Yield control so the rest of the task body can run.
-        await _real_sleep(0)
-
-    with patch.object(base_module.asyncio, "sleep", _fake_sleep):
-        adapter._schedule_ephemeral_delete(
-            chat_id="42", message_id="m-2", ttl_seconds=5
-        )
-        # Let the spawned task run.
-        for _ in range(5):
-            await _real_sleep(0)
-
-    # Only the ttl sleep shows up — the test pump uses the real sleep.
-    assert 5 in sleeps
-    assert adapter.deleted == [("42", "m-2")]
-
-
-@pytest.mark.asyncio
-async def test_schedule_ephemeral_delete_swallows_errors():
-    adapter = _delete_adapter()
-
-    async def _boom(*a, **kw):
-        raise RuntimeError("permission denied")
-
-    adapter.delete_message = _boom  # type: ignore[assignment]
-    with patch("gateway.platforms.base.asyncio.sleep", AsyncMock()):
-        adapter._schedule_ephemeral_delete(
-            chat_id="42", message_id="m-2", ttl_seconds=1
-        )
-        # No exception should propagate even though delete_message raised.
-        for _ in range(5):
-            await asyncio.sleep(0)
-
-
-def test_schedule_ephemeral_delete_outside_event_loop_is_noop():
-    """No running loop → no crash, silently drops the request."""
-    adapter = _delete_adapter()
-    # No pytest.mark.asyncio → no loop.  Must not raise.
-    adapter._schedule_ephemeral_delete(
-        chat_id="42", message_id="m-2", ttl_seconds=1
-    )
-    assert adapter.deleted == []
-
-
-# ---------------------------------------------------------------------------
-# _process_message_background unwraps EphemeralReply before send
-# ---------------------------------------------------------------------------
-
-
-@pytest.mark.asyncio
-async def test_process_message_unwraps_ephemeral_before_send():
-    """The adapter must send the wrapper's .text, never the wrapper object."""
-    adapter = _delete_adapter()
-    adapter._send_with_retry = AsyncMock(
-        return_value=SendResult(success=True, message_id="sent-1")
-    )
-
-    async def _handler(evt):
-        return EphemeralReply("⚡ Stopped.", ttl_seconds=5)
-
-    adapter.set_message_handler(_handler)
-
-    sleeps: list[float] = []
-
-    async def _fake_sleep(duration):
-        sleeps.append(duration)
-
-    event = _make_event()
-    session_key = "agent:main:telegram:private:42"
-    with patch("gateway.platforms.base.asyncio.sleep", _fake_sleep), patch.object(
-        adapter, "_keep_typing", new=AsyncMock()
-    ):
-        await adapter._process_message_background(event, session_key)
-        # Pump until the detached delete task completes.
-        for _ in range(10):
-            await asyncio.sleep(0)
-
-    # Sent text is the unwrapped string, NOT repr(EphemeralReply(...))
-    adapter._send_with_retry.assert_called_once()
-    sent_text = adapter._send_with_retry.call_args.kwargs["content"]
-    assert sent_text == "⚡ Stopped."
-    # Auto-delete scheduled using the returned message_id
-    assert ("42", "sent-1") in adapter.deleted
-
-
-@pytest.mark.asyncio
-async def test_process_message_incapable_platform_does_not_schedule_delete():
-    adapter = _no_delete_adapter()
-    adapter._send_with_retry = AsyncMock(
-        return_value=SendResult(success=True, message_id="sent-1")
-    )
-
-    async def _handler(evt):
-        return EphemeralReply("⚡ Stopped.", ttl_seconds=5)
-
-    adapter.set_message_handler(_handler)
-
-    # Spy on delete_message to confirm it is NOT invoked.
-    delete_calls: list = []
-
-    async def _spy_delete(chat_id, message_id):
-        delete_calls.append((chat_id, message_id))
-        return False
-
-    adapter.delete_message = _spy_delete  # type: ignore[assignment]
-
-    event = _make_event()
-    session_key = "agent:main:telegram:private:42"
-    with patch("gateway.platforms.base.asyncio.sleep", AsyncMock()), patch.object(
-        adapter, "_keep_typing", new=AsyncMock()
-    ):
-        await adapter._process_message_background(event, session_key)
-        for _ in range(10):
-            await asyncio.sleep(0)
-
-    # Send happened with the unwrapped text...
-    adapter._send_with_retry.assert_called_once()
-    assert adapter._send_with_retry.call_args.kwargs["content"] == "⚡ Stopped."
-    # ...but delete was never scheduled because the capability check skipped
-    # the schedule call (TTL was zeroed in _unwrap_ephemeral).
-    # Note: the capability gate on _unwrap_ephemeral checks for
-    # ``type(adapter).delete_message is BasePlatformAdapter.delete_message``.
-    # Monkeypatching the instance does NOT change the class, so this test
-    # verifies the gate uses the class method to detect capability.
-    assert delete_calls == []
-
-
-@pytest.mark.asyncio
-async def test_process_message_plain_string_behaves_unchanged():
-    adapter = _delete_adapter()
-    adapter._send_with_retry = AsyncMock(
-        return_value=SendResult(success=True, message_id="sent-1")
-    )
-
-    async def _handler(evt):
-        return "plain reply"
-
-    adapter.set_message_handler(_handler)
-
-    event = _make_event()
-    session_key = "agent:main:telegram:private:42"
-    with patch("gateway.platforms.base.asyncio.sleep", AsyncMock()), patch.object(
-        adapter, "_keep_typing", new=AsyncMock()
-    ):
-        await adapter._process_message_background(event, session_key)
-        for _ in range(5):
-            await asyncio.sleep(0)
-
-    adapter._send_with_retry.assert_called_once()
-    assert adapter._send_with_retry.call_args.kwargs["content"] == "plain reply"
-    assert adapter.deleted == []  # no auto-delete for plain replies

tests/gateway/test_feishu.py

@@ -8,7 +8,6 @@ import time
 import unittest
 from pathlib import Path
 from types import SimpleNamespace
-from typing import Dict
 from unittest.mock import AsyncMock, Mock, patch
 
 from gateway.platforms.base import ProcessingOutcome
@@ -558,16 +557,6 @@ class TestAdapterModule(unittest.TestCase):
         self.assertEqual(fake_client._ping_interval, 4)
 
 
-def _admits_group(adapter, message, sender_id, chat_id=""):
-    """Group-path shim: run a message through ``_admit`` and return a bool."""
-    sender = SimpleNamespace(sender_type="user", sender_id=sender_id)
-    if not hasattr(message, "chat_type"):
-        message.chat_type = "group"
-    if chat_id:
-        message.chat_id = chat_id
-    return adapter._admit(sender, message) is None
-
-
 class TestAdapterBehavior(unittest.TestCase):
     @patch.dict(os.environ, {}, clear=True)
     def test_build_event_handler_registers_reaction_and_card_processors(self):
@@ -700,67 +689,6 @@ class TestAdapterBehavior(unittest.TestCase):
             adapter._on_reaction_event("im.message.reaction.created_v1", data)
         run_threadsafe.assert_called_once()
 
-    def _build_reaction_adapter(self, *, msg_sender_id: str):
-        """Build a FeishuAdapter wired up to return a single GET-message result."""
-        from gateway.config import PlatformConfig
-        from gateway.platforms.feishu import FeishuAdapter
-
-        adapter = FeishuAdapter(PlatformConfig())
-        adapter._app_id = "cli_self_app"
-        adapter._bot_open_id = "ou_self_bot"
-        adapter._bot_user_id = "u_self_bot"
-
-        msg = SimpleNamespace(
-            sender=SimpleNamespace(sender_type="app", id=msg_sender_id, id_type="app_id"),
-            chat_id="oc_chat",
-            chat_type="group",
-        )
-        response = SimpleNamespace(success=lambda: True, data=SimpleNamespace(items=[msg]))
-        adapter._client = SimpleNamespace(
-            im=SimpleNamespace(
-                v1=SimpleNamespace(message=SimpleNamespace(get=Mock(return_value=response)))
-            )
-        )
-        adapter._build_get_message_request = Mock(return_value=object())
-        adapter._handle_message_with_guards = AsyncMock()
-        adapter._resolve_sender_profile = AsyncMock(
-            return_value={"user_id": "u_human", "user_name": "Human", "user_id_alt": None}
-        )
-        adapter.get_chat_info = AsyncMock(return_value={"name": "Test Chat"})
-        return adapter
-
-    @patch.dict(os.environ, {}, clear=True)
-    def test_reaction_on_peer_bot_message_is_not_routed(self):
-        # GET im/v1/messages sender for bot messages carries id=app_id; a peer
-        # bot's message has a different app_id than ours, so it must be dropped.
-        adapter = self._build_reaction_adapter(msg_sender_id="cli_peer_app")
-
-        event = SimpleNamespace(
-            message_id="om_peer_msg",
-            user_id=SimpleNamespace(open_id="ou_human", user_id=None, union_id=None),
-            reaction_type=SimpleNamespace(emoji_type="THUMBSUP"),
-        )
-        data = SimpleNamespace(event=event)
-        asyncio.run(
-            adapter._handle_reaction_event("im.message.reaction.created_v1", data)
-        )
-        adapter._handle_message_with_guards.assert_not_awaited()
-
-    @patch.dict(os.environ, {}, clear=True)
-    def test_reaction_on_our_own_bot_message_is_routed(self):
-        adapter = self._build_reaction_adapter(msg_sender_id="cli_self_app")
-
-        event = SimpleNamespace(
-            message_id="om_self_msg",
-            user_id=SimpleNamespace(open_id="ou_human", user_id=None, union_id=None),
-            reaction_type=SimpleNamespace(emoji_type="THUMBSUP"),
-        )
-        data = SimpleNamespace(event=event)
-        asyncio.run(
-            adapter._handle_reaction_event("im.message.reaction.created_v1", data)
-        )
-        adapter._handle_message_with_guards.assert_awaited_once()
-
     @patch.dict(os.environ, {"FEISHU_GROUP_POLICY": "open"}, clear=True)
     def test_group_message_requires_mentions_even_when_policy_open(self):
         from gateway.config import PlatformConfig
@@ -769,10 +697,10 @@ class TestAdapterBehavior(unittest.TestCase):
         adapter = FeishuAdapter(PlatformConfig())
         message = SimpleNamespace(mentions=[])
         sender_id = SimpleNamespace(open_id="ou_any", user_id=None)
-        self.assertFalse(_admits_group(adapter, message, sender_id, ""))
+        self.assertFalse(adapter._should_accept_group_message(message, sender_id, ""))
 
         message_with_mention = SimpleNamespace(mentions=[SimpleNamespace(key="@_user_1")])
-        self.assertFalse(_admits_group(adapter, message_with_mention, sender_id, ""))
+        self.assertFalse(adapter._should_accept_group_message(message_with_mention, sender_id, ""))
 
     @patch.dict(os.environ, {"FEISHU_GROUP_POLICY": "open"}, clear=True)
     def test_group_message_with_other_user_mention_is_rejected_when_bot_identity_unknown(self):
@@ -786,10 +714,59 @@ class TestAdapterBehavior(unittest.TestCase):
             id=SimpleNamespace(open_id="ou_other", user_id="u_other"),
         )
 
-        self.assertFalse(
-            _admits_group(adapter, SimpleNamespace(mentions=[other_mention]), sender_id, "")
+        self.assertFalse(adapter._should_accept_group_message(SimpleNamespace(mentions=[other_mention]), sender_id, ""))
+
+    @patch.dict(
+        os.environ,
+        {
+            "FEISHU_BOT_OPEN_ID": "ou_hermes",
+            "FEISHU_BOT_USER_ID": "u_hermes",
+        },
+        clear=True,
+    )
+    def test_other_bot_sender_is_not_treated_as_self_sent_message(self):
+        from gateway.config import PlatformConfig
+        from gateway.platforms.feishu import FeishuAdapter
+
+        adapter = FeishuAdapter(PlatformConfig())
+        event = SimpleNamespace(
+            sender=SimpleNamespace(
+                sender_type="bot",
+                sender_id=SimpleNamespace(open_id="ou_other_bot", user_id="u_other_bot"),
+            )
         )
 
+        self.assertFalse(adapter._is_self_sent_bot_message(event))
+
+    @patch.dict(
+        os.environ,
+        {
+            "FEISHU_BOT_OPEN_ID": "ou_hermes",
+            "FEISHU_BOT_USER_ID": "u_hermes",
+        },
+        clear=True,
+    )
+    def test_self_bot_sender_is_treated_as_self_sent_message(self):
+        from gateway.config import PlatformConfig
+        from gateway.platforms.feishu import FeishuAdapter
+
+        adapter = FeishuAdapter(PlatformConfig())
+        by_open_id = SimpleNamespace(
+            sender=SimpleNamespace(
+                sender_type="bot",
+                sender_id=SimpleNamespace(open_id="ou_hermes", user_id="u_other"),
+            )
+        )
+        by_user_id = SimpleNamespace(
+            sender=SimpleNamespace(
+                sender_type="app",
+                sender_id=SimpleNamespace(open_id="ou_other", user_id="u_hermes"),
+            )
+        )
+
+        self.assertTrue(adapter._is_self_sent_bot_message(by_open_id))
+        self.assertTrue(adapter._is_self_sent_bot_message(by_user_id))
+
     @patch.dict(
         os.environ,
         {
@@ -815,14 +792,14 @@ class TestAdapterBehavior(unittest.TestCase):
         )
 
         self.assertTrue(
-            _admits_group(adapter,
+            adapter._should_accept_group_message(
                 mentioned,
                 SimpleNamespace(open_id="ou_allowed", user_id=None),
                 "",
             )
         )
         self.assertFalse(
-            _admits_group(adapter,
+            adapter._should_accept_group_message(
                 mentioned,
                 SimpleNamespace(open_id="ou_blocked", user_id=None),
                 "",
@@ -851,14 +828,14 @@ class TestAdapterBehavior(unittest.TestCase):
         )
 
         self.assertTrue(
-            _admits_group(adapter,
+            adapter._should_accept_group_message(
                 message,
                 SimpleNamespace(open_id="ou_alice", user_id=None),
                 "oc_chat_a",
             )
         )
         self.assertFalse(
-            _admits_group(adapter,
+            adapter._should_accept_group_message(
                 message,
                 SimpleNamespace(open_id="ou_charlie", user_id=None),
                 "oc_chat_a",
@@ -887,14 +864,14 @@ class TestAdapterBehavior(unittest.TestCase):
         )
 
         self.assertTrue(
-            _admits_group(adapter,
+            adapter._should_accept_group_message(
                 message,
                 SimpleNamespace(open_id="ou_alice", user_id=None),
                 "oc_chat_b",
             )
         )
         self.assertFalse(
-            _admits_group(adapter,
+            adapter._should_accept_group_message(
                 message,
                 SimpleNamespace(open_id="ou_blocked", user_id=None),
                 "oc_chat_b",
@@ -923,14 +900,14 @@ class TestAdapterBehavior(unittest.TestCase):
         )
 
         self.assertTrue(
-            _admits_group(adapter,
+            adapter._should_accept_group_message(
                 message,
                 SimpleNamespace(open_id="ou_admin", user_id=None),
                 "oc_chat_c",
             )
         )
         self.assertFalse(
-            _admits_group(adapter,
+            adapter._should_accept_group_message(
                 message,
                 SimpleNamespace(open_id="ou_regular", user_id=None),
                 "oc_chat_c",
@@ -959,14 +936,14 @@ class TestAdapterBehavior(unittest.TestCase):
         )
 
         self.assertTrue(
-            _admits_group(adapter,
+            adapter._should_accept_group_message(
                 message,
                 SimpleNamespace(open_id="ou_admin", user_id=None),
                 "oc_chat_d",
             )
         )
         self.assertFalse(
-            _admits_group(adapter,
+            adapter._should_accept_group_message(
                 message,
                 SimpleNamespace(open_id="ou_regular", user_id=None),
                 "oc_chat_d",
@@ -996,7 +973,7 @@ class TestAdapterBehavior(unittest.TestCase):
         )
 
         self.assertTrue(
-            _admits_group(adapter,
+            adapter._should_accept_group_message(
                 message,
                 SimpleNamespace(open_id="ou_admin", user_id=None),
                 "oc_chat_e",
@@ -1020,7 +997,7 @@ class TestAdapterBehavior(unittest.TestCase):
         )
 
         self.assertTrue(
-            _admits_group(adapter,
+            adapter._should_accept_group_message(
                 message,
                 SimpleNamespace(open_id="ou_anyone", user_id=None),
                 "oc_chat_unknown",
@@ -1045,12 +1022,8 @@ class TestAdapterBehavior(unittest.TestCase):
             id=SimpleNamespace(open_id="ou_other", user_id="u_other"),
         )
 
-        self.assertTrue(
-            _admits_group(adapter, SimpleNamespace(mentions=[bot_mention]), sender_id, "")
-        )
-        self.assertFalse(
-            _admits_group(adapter, SimpleNamespace(mentions=[other_mention]), sender_id, "")
-        )
+        self.assertTrue(adapter._should_accept_group_message(SimpleNamespace(mentions=[bot_mention]), sender_id, ""))
+        self.assertFalse(adapter._should_accept_group_message(SimpleNamespace(mentions=[other_mention]), sender_id, ""))
 
     @patch.dict(os.environ, {"FEISHU_GROUP_POLICY": "open"}, clear=True)
     def test_group_message_matches_bot_name_when_only_name_available(self):
@@ -1075,12 +1048,8 @@ class TestAdapterBehavior(unittest.TestCase):
             id=SimpleNamespace(open_id=None, user_id=None),
         )
 
-        self.assertTrue(
-            _admits_group(adapter, SimpleNamespace(mentions=[name_only_mention]), sender_id, "")
-        )
-        self.assertFalse(
-            _admits_group(adapter, SimpleNamespace(mentions=[different_mention]), sender_id, "")
-        )
+        self.assertTrue(adapter._should_accept_group_message(SimpleNamespace(mentions=[name_only_mention]), sender_id, ""))
+        self.assertFalse(adapter._should_accept_group_message(SimpleNamespace(mentions=[different_mention]), sender_id, ""))
 
         # Case 2: bot's open_id IS known — a same-name human with different
         # open_id must NOT admit (IDs override names).
@@ -1097,17 +1066,8 @@ class TestAdapterBehavior(unittest.TestCase):
             id=SimpleNamespace(open_id="ou_bot", user_id=None),
         )
 
-        self.assertFalse(
-            _admits_group(
-                adapter2,
-                SimpleNamespace(mentions=[same_name_other_id_mention]),
-                sender_id,
-                "",
-            )
-        )
-        self.assertTrue(
-            _admits_group(adapter2, SimpleNamespace(mentions=[bot_mention]), sender_id, "")
-        )
+        self.assertFalse(adapter2._should_accept_group_message(SimpleNamespace(mentions=[same_name_other_id_mention]), sender_id, ""))
+        self.assertTrue(adapter2._should_accept_group_message(SimpleNamespace(mentions=[bot_mention]), sender_id, ""))
 
     @patch.dict(os.environ, {}, clear=True)
     def test_extract_post_message_as_text(self):
@@ -1451,7 +1411,6 @@ class TestAdapterBehavior(unittest.TestCase):
                 data=SimpleNamespace(event=SimpleNamespace(message=message)),
                 message=message,
                 sender_id=SimpleNamespace(open_id="ou_user", user_id=None, union_id=None),
-                is_bot=False,
                 chat_type="p2p",
                 message_id="om_command",
             )
@@ -1563,14 +1522,13 @@ class TestAdapterBehavior(unittest.TestCase):
             user_id="u_user",
             union_id="on_union",
         )
-        sender = SimpleNamespace(sender_type="user", sender_id=sender_id)
-        data = SimpleNamespace(event=SimpleNamespace(message=message, sender=sender))
+        data = SimpleNamespace(event=SimpleNamespace(message=message, sender=SimpleNamespace(sender_id=sender_id)))
 
         asyncio.run(
             adapter._process_inbound_message(
                 data=data,
                 message=message,
-                sender_id=sender.sender_id,
+                sender_id=sender_id,
                 chat_type="p2p",
                 message_id="om_text",
             )
@@ -1803,14 +1761,13 @@ class TestAdapterBehavior(unittest.TestCase):
             message_id="om_group_text",
         )
         sender_id = SimpleNamespace(open_id="ou_user", user_id=None, union_id=None)
-        sender = SimpleNamespace(sender_type="user", sender_id=sender_id)
         data = SimpleNamespace(event=SimpleNamespace(message=message))
 
         asyncio.run(
             adapter._process_inbound_message(
                 data=data,
                 message=message,
-                sender_id=sender.sender_id,
+                sender_id=sender_id,
                 chat_type="group",
                 message_id="om_group_text",
             )
@@ -1848,7 +1805,6 @@ class TestAdapterBehavior(unittest.TestCase):
                 data=SimpleNamespace(event=SimpleNamespace(message=message)),
                 message=message,
                 sender_id=SimpleNamespace(open_id="ou_user", user_id=None, union_id=None),
-                is_bot=False,
                 chat_type="p2p",
                 message_id="om_reply",
             )
@@ -2711,12 +2667,11 @@ class TestAdapterBehavior(unittest.TestCase):
 
 @unittest.skipUnless(_HAS_LARK_OAPI, "lark-oapi not installed")
 class TestHydrateBotIdentity(unittest.TestCase):
-    """Hydration of bot identity via ``/open-apis/bot/v3/info``.
+    """Hydration of bot identity via /open-apis/bot/v3/info and application info.
 
-    Covers the manual-setup path where ``FEISHU_BOT_OPEN_ID`` /
-    ``FEISHU_BOT_NAME`` are not configured — hydration populates them so
-    self-echo protection and group @mention gating both have something to
-    match against.
+    Covers the manual-setup path where FEISHU_BOT_OPEN_ID / FEISHU_BOT_USER_ID
+    are not configured. Hydration must populate _bot_open_id so that
+    _is_self_sent_bot_message() can filter the adapter's own outbound echoes.
     """
 
     def _make_adapter(self):
@@ -2745,6 +2700,11 @@ class TestHydrateBotIdentity(unittest.TestCase):
 
         self.assertEqual(adapter._bot_open_id, "ou_hermes_hydrated")
         self.assertEqual(adapter._bot_name, "Hermes Bot")
+        # Application-info fallback must NOT run when bot_name is already set.
+        self.assertFalse(
+            adapter._client.application.v6.application.get.called
+            if hasattr(adapter._client, "application") else False
+        )
 
     @patch.dict(
         os.environ,
@@ -2761,6 +2721,7 @@ class TestHydrateBotIdentity(unittest.TestCase):
 
         asyncio.run(adapter._hydrate_bot_identity())
 
+        # Neither probe should run — both fields are already populated.
         adapter._client.request.assert_not_called()
         self.assertEqual(adapter._bot_open_id, "ou_env")
         self.assertEqual(adapter._bot_name, "Env Hermes")
@@ -2805,6 +2766,33 @@ class TestHydrateBotIdentity(unittest.TestCase):
         self.assertEqual(adapter._bot_open_id, "")
         self.assertEqual(adapter._bot_name, "Fallback Bot")
 
+    @patch.dict(os.environ, {}, clear=True)
+    def test_hydrated_open_id_enables_self_send_filter(self):
+        """E2E: after hydration, _is_self_sent_bot_message() rejects adapter's own id."""
+        adapter = self._make_adapter()
+        adapter._client = Mock()
+        payload = json.dumps(
+            {"code": 0, "bot": {"bot_name": "Hermes", "open_id": "ou_hermes"}}
+        ).encode("utf-8")
+        adapter._client.request = Mock(return_value=SimpleNamespace(raw=SimpleNamespace(content=payload)))
+
+        asyncio.run(adapter._hydrate_bot_identity())
+
+        self_event = SimpleNamespace(
+            sender=SimpleNamespace(
+                sender_type="bot",
+                sender_id=SimpleNamespace(open_id="ou_hermes", user_id=""),
+            )
+        )
+        peer_event = SimpleNamespace(
+            sender=SimpleNamespace(
+                sender_type="bot",
+                sender_id=SimpleNamespace(open_id="ou_peer_bot", user_id=""),
+            )
+        )
+        self.assertTrue(adapter._is_self_sent_bot_message(self_event))
+        self.assertFalse(adapter._is_self_sent_bot_message(peer_event))
+
 
 @unittest.skipUnless(_HAS_LARK_OAPI, "lark-oapi not installed")
 class TestPendingInboundQueue(unittest.TestCase):
@@ -3149,7 +3137,7 @@ class TestGroupMentionAtAll(unittest.TestCase):
             mentions=[],
         )
         sender_id = SimpleNamespace(open_id="ou_any", user_id=None)
-        self.assertTrue(_admits_group(adapter, message, sender_id, ""))
+        self.assertTrue(adapter._should_accept_group_message(message, sender_id, ""))
 
     @patch.dict(os.environ, {"FEISHU_GROUP_POLICY": "allowlist", "FEISHU_ALLOWED_USERS": "ou_allowed"}, clear=True)
     def test_at_all_still_requires_policy_gate(self):
@@ -3161,15 +3149,15 @@ class TestGroupMentionAtAll(unittest.TestCase):
         message = SimpleNamespace(content='{"text":"@_all attention"}', mentions=[])
         # Non-allowlisted user — should be blocked even with @_all.
         blocked_sender = SimpleNamespace(open_id="ou_blocked", user_id=None)
-        self.assertFalse(_admits_group(adapter, message, blocked_sender, ""))
+        self.assertFalse(adapter._should_accept_group_message(message, blocked_sender, ""))
         # Allowlisted user — should pass.
         allowed_sender = SimpleNamespace(open_id="ou_allowed", user_id=None)
-        self.assertTrue(_admits_group(adapter, message, allowed_sender, ""))
+        self.assertTrue(adapter._should_accept_group_message(message, allowed_sender, ""))
 
 
 @unittest.skipUnless(_HAS_LARK_OAPI, "lark-oapi not installed")
 class TestSenderNameResolution(unittest.TestCase):
-    """Tests for _resolve_sender_name_from_api (contact API + cache)."""
+    """Tests for _resolve_sender_name_from_api."""
 
     @patch.dict(os.environ, {}, clear=True)
     def test_returns_none_when_client_is_none(self):
@@ -3273,137 +3261,6 @@ class TestSenderNameResolution(unittest.TestCase):
         self.assertIsNone(result)
 
 
-@unittest.skipUnless(_HAS_LARK_OAPI, "lark-oapi not installed")
-class TestBotNameResolution(unittest.TestCase):
-    """Tests for the bot branch of _resolve_sender_name_from_api (basic_batch API + shared cache)."""
-
-    @staticmethod
-    def _batch_payload(bots: Dict[str, str]):
-        import json as _json
-        body = {
-            oid: {"bot_id": oid, "name": name, "i18n_names": {"en_us": name}}
-            for oid, name in bots.items()
-        }
-        return _json.dumps({"code": 0, "msg": "", "data": {"bots": body, "failed_bots": {}}}).encode()
-
-    def _build_adapter_with_bots(self, bots: Dict[str, str]):
-        from gateway.config import PlatformConfig
-        from gateway.platforms.feishu import FeishuAdapter
-
-        adapter = FeishuAdapter(PlatformConfig())
-        calls = []
-
-        def _fake_request(request):
-            calls.append(request)
-            return SimpleNamespace(raw=SimpleNamespace(content=self._batch_payload(bots)))
-
-        adapter._client = SimpleNamespace(request=_fake_request)
-        return adapter, calls
-
-    @patch.dict(os.environ, {}, clear=True)
-    def test_returns_cached_bot_name_without_api_call(self):
-        from gateway.config import PlatformConfig
-        from gateway.platforms.feishu import FeishuAdapter
-
-        adapter = FeishuAdapter(PlatformConfig())
-        adapter._sender_name_cache["ou_peer"] = ("Peer Bot", time.time() + 600)
-        adapter._client = SimpleNamespace(
-            request=lambda _r: (_ for _ in ()).throw(RuntimeError("should not fetch"))
-        )
-        result = asyncio.run(adapter._resolve_sender_name_from_api("ou_peer", is_bot=True))
-        self.assertEqual(result, "Peer Bot")
-
-    @patch.dict(os.environ, {}, clear=True)
-    def test_fetches_and_caches_bot_name(self):
-        adapter, calls = self._build_adapter_with_bots({"ou_peer": "Peer Bot"})
-
-        async def _direct(func, *args, **kwargs):
-            return func(*args, **kwargs)
-
-        with patch("gateway.platforms.feishu.asyncio.to_thread", side_effect=_direct):
-            result = asyncio.run(adapter._resolve_sender_name_from_api("ou_peer", is_bot=True))
-
-        self.assertEqual(result, "Peer Bot")
-        self.assertEqual(adapter._sender_name_cache["ou_peer"][0], "Peer Bot")
-        self.assertEqual(len(calls), 1)
-        self.assertIn("/open-apis/bot/v3/bots/basic_batch", calls[0].uri)
-        # Feishu expects repeated ?bot_ids= params, not comma-joined.
-        self.assertEqual(calls[0].queries, [("bot_ids", "ou_peer")])
-
-    @patch.dict(os.environ, {}, clear=True)
-    def test_api_failure_returns_none_and_does_not_poison_cache(self):
-        from gateway.config import PlatformConfig
-        from gateway.platforms.feishu import FeishuAdapter
-
-        adapter = FeishuAdapter(PlatformConfig())
-
-        def _broken_request(_req):
-            raise RuntimeError("API down")
-
-        adapter._client = SimpleNamespace(request=_broken_request)
-
-        async def _direct(func, *args, **kwargs):
-            return func(*args, **kwargs)
-
-        with patch("gateway.platforms.feishu.asyncio.to_thread", side_effect=_direct):
-            result = asyncio.run(adapter._resolve_sender_name_from_api("ou_peer", is_bot=True))
-
-        self.assertIsNone(result)
-        self.assertNotIn("ou_peer", adapter._sender_name_cache)
-
-    @patch.dict(os.environ, {}, clear=True)
-    def test_bot_absent_from_response_is_not_cached(self):
-        """Bot not in ``data.bots`` (e.g. landed in ``failed_bots``) → no
-        cache entry, next lookup re-fetches."""
-        adapter, _ = self._build_adapter_with_bots({"ou_other": "Other Bot"})
-
-        async def _direct(func, *args, **kwargs):
-            return func(*args, **kwargs)
-
-        with patch("gateway.platforms.feishu.asyncio.to_thread", side_effect=_direct):
-            result = asyncio.run(adapter._resolve_sender_name_from_api("ou_ghost", is_bot=True))
-
-        self.assertIsNone(result)
-        self.assertNotIn("ou_ghost", adapter._sender_name_cache)
-
-    @patch.dict(os.environ, {}, clear=True)
-    def test_empty_name_in_response_is_negative_cached(self):
-        """API returns name="" → cache "" so repeat lookups short-circuit."""
-        adapter, calls = self._build_adapter_with_bots({"ou_nameless": ""})
-
-        async def _direct(func, *args, **kwargs):
-            return func(*args, **kwargs)
-
-        with patch("gateway.platforms.feishu.asyncio.to_thread", side_effect=_direct):
-            first = asyncio.run(adapter._resolve_sender_name_from_api("ou_nameless", is_bot=True))
-            second = asyncio.run(adapter._resolve_sender_name_from_api("ou_nameless", is_bot=True))
-
-        self.assertIsNone(first)
-        self.assertIsNone(second)
-        self.assertEqual(adapter._sender_name_cache["ou_nameless"][0], "")
-        self.assertEqual(len(calls), 1)
-
-    @patch.dict(os.environ, {}, clear=True)
-    def test_non_zero_code_returns_none(self):
-        from gateway.config import PlatformConfig
-        from gateway.platforms.feishu import FeishuAdapter
-
-        adapter = FeishuAdapter(PlatformConfig())
-        error_payload = b'{"code":99991663,"msg":"permission denied"}'
-        adapter._client = SimpleNamespace(
-            request=lambda _r: SimpleNamespace(raw=SimpleNamespace(content=error_payload))
-        )
-
-        async def _direct(func, *args, **kwargs):
-            return func(*args, **kwargs)
-
-        with patch("gateway.platforms.feishu.asyncio.to_thread", side_effect=_direct):
-            result = asyncio.run(adapter._resolve_sender_name_from_api("ou_peer", is_bot=True))
-
-        self.assertIsNone(result)
-        self.assertNotIn("ou_peer", adapter._sender_name_cache)
-
-
 @unittest.skipUnless(_HAS_LARK_OAPI, "lark-oapi not installed")
 class TestProcessingReactions(unittest.TestCase):
     """Typing on start → removed on SUCCESS, swapped for CrossMark on FAILURE,

tests/gateway/test_feishu_bot_admission.py

@@ -1,745 +0,0 @@
-"""Adapter-layer tests for Feishu bot-sender admission (``FeishuAdapter._admit``)."""
-
-from __future__ import annotations
-
-from types import SimpleNamespace
-from typing import Any
-
-import pytest
-
-from tests.gateway.feishu_helpers import (
-    install_dedup_state,
-    make_adapter_skeleton,
-    make_message,
-    make_sender,
-    stub_mention,
-)
-
-
-# --- FeishuAdapterSettings wiring ------------------------------------------
-
-
-@pytest.mark.parametrize(
-    "env_value, expected",
-    [
-        ("none", "none"),
-        ("mentions", "mentions"),
-        ("all", "all"),
-        ("  Mentions  ", "mentions"),
-    ],
-)
-def test_feishu_load_settings_populates_allow_bots(monkeypatch, env_value, expected):
-    from gateway.platforms.feishu import FeishuAdapter
-
-    monkeypatch.setenv("FEISHU_APP_ID", "cli_test")
-    monkeypatch.setenv("FEISHU_APP_SECRET", "secret_test")
-    monkeypatch.setenv("FEISHU_ALLOW_BOTS", env_value)
-
-    settings = FeishuAdapter._load_settings(extra={})
-    assert settings.allow_bots == expected
-
-
-def test_feishu_load_settings_allow_bots_defaults_to_none(monkeypatch):
-    from gateway.platforms.feishu import FeishuAdapter
-
-    monkeypatch.setenv("FEISHU_APP_ID", "cli_test")
-    monkeypatch.setenv("FEISHU_APP_SECRET", "secret_test")
-    monkeypatch.delenv("FEISHU_ALLOW_BOTS", raising=False)
-
-    settings = FeishuAdapter._load_settings(extra={})
-    assert settings.allow_bots == "none"
-
-
-def test_feishu_load_settings_ignores_extra_allow_bots(monkeypatch):
-    # extra is ignored — env is single source of truth (yaml is bridged to env).
-    from gateway.platforms.feishu import FeishuAdapter
-
-    monkeypatch.setenv("FEISHU_APP_ID", "cli_test")
-    monkeypatch.setenv("FEISHU_APP_SECRET", "secret_test")
-    monkeypatch.delenv("FEISHU_ALLOW_BOTS", raising=False)
-
-    settings = FeishuAdapter._load_settings(extra={"allow_bots": "all"})
-    assert settings.allow_bots == "none"
-
-
-def test_feishu_load_settings_falls_back_to_env_when_extra_missing(monkeypatch):
-    from gateway.platforms.feishu import FeishuAdapter
-
-    monkeypatch.setenv("FEISHU_APP_ID", "cli_test")
-    monkeypatch.setenv("FEISHU_APP_SECRET", "secret_test")
-    monkeypatch.setenv("FEISHU_ALLOW_BOTS", "mentions")
-
-    settings = FeishuAdapter._load_settings(extra={})
-    assert settings.allow_bots == "mentions"
-
-
-def test_feishu_load_settings_warns_on_unknown_allow_bots(monkeypatch, caplog):
-    import logging
-
-    from gateway.platforms.feishu import FeishuAdapter
-
-    monkeypatch.setenv("FEISHU_APP_ID", "cli_test")
-    monkeypatch.setenv("FEISHU_APP_SECRET", "secret_test")
-    monkeypatch.setenv("FEISHU_ALLOW_BOTS", "menton")  # typo
-
-    with caplog.at_level(logging.WARNING, logger="gateway.platforms.feishu"):
-        settings = FeishuAdapter._load_settings(extra={})
-
-    assert settings.allow_bots == "none"
-    assert any("allow_bots" in r.message and "menton" in r.message for r in caplog.records)
-
-
-@pytest.mark.parametrize(
-    "env_value, extra, expected",
-    [
-        (None, {}, True),
-        ("false", {}, False),
-        ("true", {}, True),
-        ("true", {"require_mention": False}, False),
-    ],
-)
-def test_feishu_load_settings_require_mention(monkeypatch, env_value, extra, expected):
-    from gateway.platforms.feishu import FeishuAdapter
-
-    monkeypatch.setenv("FEISHU_APP_ID", "cli_test")
-    monkeypatch.setenv("FEISHU_APP_SECRET", "secret_test")
-    if env_value is None:
-        monkeypatch.delenv("FEISHU_REQUIRE_MENTION", raising=False)
-    else:
-        monkeypatch.setenv("FEISHU_REQUIRE_MENTION", env_value)
-
-    settings = FeishuAdapter._load_settings(extra=extra)
-    assert settings.require_mention is expected
-
-
-def test_feishu_load_settings_parses_per_group_require_mention(monkeypatch):
-    from gateway.platforms.feishu import FeishuAdapter
-
-    monkeypatch.setenv("FEISHU_APP_ID", "cli_test")
-    monkeypatch.setenv("FEISHU_APP_SECRET", "secret_test")
-
-    settings = FeishuAdapter._load_settings(extra={
-        "group_rules": {
-            "oc_free": {"policy": "open", "require_mention": False},
-            "oc_strict": {"policy": "open", "require_mention": True},
-            "oc_inherit": {"policy": "open"},
-        },
-    })
-    assert settings.group_rules["oc_free"].require_mention is False
-    assert settings.group_rules["oc_strict"].require_mention is True
-    assert settings.group_rules["oc_inherit"].require_mention is None
-
-
-# --- Module-level helpers --------------------------------------------------
-
-
-def test_sender_identity_collects_every_non_empty_id_variant():
-    from gateway.platforms.feishu import _sender_identity
-
-    sender = SimpleNamespace(
-        sender_id=SimpleNamespace(open_id="ou_x", user_id="", union_id="un_x"),
-    )
-    assert _sender_identity(sender) == frozenset({"ou_x", "un_x"})
-
-
-def test_sender_identity_handles_missing_sender_id():
-    from gateway.platforms.feishu import _sender_identity
-
-    assert _sender_identity(SimpleNamespace()) == frozenset()
-
-
-@pytest.mark.parametrize("sender_type", ["bot", "app"])
-def test_is_bot_sender_treats_bot_and_app_as_bot_origin(sender_type):
-    from gateway.platforms.feishu import _is_bot_sender
-
-    assert _is_bot_sender(SimpleNamespace(sender_type=sender_type)) is True
-
-
-@pytest.mark.parametrize("sender_type", ["user", "", None])
-def test_is_bot_sender_rejects_non_bot_origin(sender_type):
-    from gateway.platforms.feishu import _is_bot_sender
-
-    assert _is_bot_sender(SimpleNamespace(sender_type=sender_type)) is False
-
-
-# --- _admit pipeline matrix ------------------------------------------------
-#
-# Covers the four-step admission pipeline (self_echo → bot_policy →
-# DM bypass → group_policy + mention) as a single result-only matrix.
-# Each row pins one decision in the pipeline; tests asserting call-count
-# semantics live below in their own functions.
-
-
-def _admit_case(
-    *,
-    adapter: dict | None = None,
-    sender: dict | None = None,
-    message: dict | None = None,
-    mentions_self: bool | None = None,
-    expected: str | None = None,
-):
-    return {
-        "adapter": adapter or {},
-        "sender": sender or {},
-        "message": message or {},
-        "mentions_self": mentions_self,
-        "expected": expected,
-    }
-
-
-_ADMIT_CASES = [
-    pytest.param(
-        _admit_case(
-            adapter={"bot_open_id": "ou_me", "allow_bots": "all"},
-            sender={"sender_type": "bot", "open_id": "ou_me"},
-            expected="self_echo",
-        ),
-        id="self_echo:open_id_under_all_mode",
-    ),
-    pytest.param(
-        _admit_case(
-            adapter={"bot_open_id": "", "bot_user_id": "u_me", "allow_bots": "all"},
-            sender={"sender_type": "bot", "open_id": None, "user_id": "u_me"},
-            expected="self_echo",
-        ),
-        id="self_echo:user_id_only",
-    ),
-    pytest.param(
-        _admit_case(
-            adapter={"bot_open_id": "ou_me", "allow_bots": "all"},
-            sender={"sender_type": "bot", "open_id": "ou_me", "user_id": "u_me", "union_id": "un_me"},
-            expected="self_echo",
-        ),
-        id="self_echo:mixed_ids",
-    ),
-    pytest.param(
-        _admit_case(
-            adapter={"bot_open_id": "ou_self", "bot_user_id": "u_self", "allow_bots": "all"},
-            sender={"sender_type": "bot", "open_id": None, "user_id": "u_self"},
-            expected="self_echo",
-        ),
-        id="self_echo:user_id_when_bot_user_id_set",
-    ),
-    pytest.param(
-        _admit_case(
-            adapter={"bot_open_id": "ou_self", "allow_bots": "none"},
-            sender={"sender_type": "bot", "open_id": "ou_peer"},
-            expected="bots_disabled",
-        ),
-        id="bots_disabled:mode_none",
-    ),
-    pytest.param(
-        _admit_case(
-            adapter={"bot_open_id": "ou_self", "allow_bots": ""},
-            sender={"sender_type": "bot", "open_id": "ou_peer"},
-            expected="bots_disabled",
-        ),
-        id="bots_disabled:mode_empty",
-    ),
-    pytest.param(
-        _admit_case(
-            adapter={"bot_open_id": "ou_self", "allow_bots": "loose"},
-            sender={"sender_type": "bot", "open_id": "ou_peer"},
-            expected="bots_disabled",
-        ),
-        id="bots_disabled:mode_unknown_value",
-    ),
-    pytest.param(
-        _admit_case(
-            adapter={"bot_open_id": "", "allow_bots": "none"},
-            sender={"sender_type": "bot", "open_id": "ou_peer"},
-            expected="bots_disabled",
-        ),
-        id="bots_disabled:wins_over_self_ids_unknown",
-    ),
-    pytest.param(
-        _admit_case(
-            adapter={"bot_open_id": "", "allow_bots": "all"},
-            sender={"sender_type": "bot", "open_id": "ou_peer"},
-            expected="self_ids_unknown",
-        ),
-        id="self_ids_unknown:bot_sender_no_self_ids",
-    ),
-    pytest.param(
-        _admit_case(
-            adapter={"bot_open_id": "", "allow_bots": "all"},
-            sender={"sender_type": "app", "open_id": "ou_peer"},
-            expected="self_ids_unknown",
-        ),
-        id="self_ids_unknown:app_sender_no_self_ids",
-    ),
-    pytest.param(
-        _admit_case(
-            adapter={"bot_open_id": "ou_self", "allow_bots": "all"},
-            sender={"sender_type": "app", "open_id": None},
-            expected="self_ids_unknown",
-        ),
-        id="self_ids_unknown:no_sender_ids",
-    ),
-    pytest.param(
-        _admit_case(
-            adapter={"bot_open_id": "ou_self", "allow_bots": "mentions"},
-            sender={"sender_type": "bot", "open_id": "ou_peer"},
-            mentions_self=False,
-            expected="bot_not_mentioned",
-        ),
-        id="mentions_mode:not_mentioned_dm",
-    ),
-    pytest.param(
-        _admit_case(
-            adapter={"bot_open_id": "ou_self", "allow_bots": "mentions"},
-            sender={"sender_type": "bot", "open_id": "ou_peer"},
-            mentions_self=True,
-            expected=None,
-        ),
-        id="mentions_mode:mentioned_dm",
-    ),
-    pytest.param(
-        _admit_case(
-            adapter={"bot_open_id": "ou_self", "allow_bots": "all"},
-            sender={"sender_type": "bot", "open_id": "ou_peer"},
-            mentions_self=False,
-            expected=None,
-        ),
-        id="all_mode:not_mentioned_dm",
-    ),
-    pytest.param(
-        _admit_case(
-            adapter={"bot_open_id": "ou_self", "allow_bots": "all"},
-            sender={"sender_type": "bot", "open_id": "ou_peer"},
-            mentions_self=True,
-            expected=None,
-        ),
-        id="all_mode:mentioned_dm",
-    ),
-    pytest.param(
-        _admit_case(
-            adapter={"bot_open_id": "", "allow_bots": "none"},
-            sender={"sender_type": "user", "open_id": "ou_human"},
-            expected=None,
-        ),
-        id="human:dm_admitted_regardless_of_allow_bots",
-    ),
-    pytest.param(
-        _admit_case(
-            adapter={"allow_bots": "all"},
-            sender={"sender_type": "user", "open_id": "ou_human"},
-            message={"message_id": "om_ok", "chat_type": "p2p"},
-            expected=None,
-        ),
-        id="human:p2p_admitted",
-    ),
-    pytest.param(
-        _admit_case(
-            adapter={
-                "bot_open_id": "ou_self",
-                "require_mention": False,
-                "group_policy": "open",
-            },
-            sender={"sender_type": "user", "open_id": "ou_human"},
-            message={"chat_type": "group"},
-            mentions_self=False,
-            expected=None,
-        ),
-        id="require_mention_false:group_human_no_mention_admitted",
-    ),
-    pytest.param(
-        _admit_case(
-            adapter={
-                "bot_open_id": "ou_self",
-                "allow_bots": "all",
-                "require_mention": False,
-                "group_policy": "open",
-            },
-            sender={"sender_type": "bot", "open_id": "ou_peer"},
-            message={"chat_type": "group"},
-            mentions_self=False,
-            expected=None,
-        ),
-        id="require_mention_false:group_bot_all_mode_admitted",
-    ),
-    pytest.param(
-        _admit_case(
-            adapter={
-                "bot_open_id": "ou_self",
-                "allow_bots": "mentions",
-                "require_mention": False,
-                "group_policy": "open",
-            },
-            sender={"sender_type": "bot", "open_id": "ou_peer"},
-            message={"chat_type": "group"},
-            mentions_self=False,
-            expected="bot_not_mentioned",
-        ),
-        id="require_mention_false:group_bot_mentions_mode_still_gated",
-    ),
-]
-
-
-@pytest.mark.parametrize("case", _ADMIT_CASES)
-def test_admit_pipeline(case):
-    adapter = make_adapter_skeleton(**case["adapter"])
-    if case["mentions_self"] is not None:
-        stub_mention(adapter, case["mentions_self"])
-    sender = make_sender(**case["sender"])
-    message = make_message(**case["message"])
-    assert adapter._admit(sender, message) == case["expected"]
-
-
-# --- Mention call-count semantics ------------------------------------------
-
-
-def test_admit_skips_mention_check_under_all_mode():
-    # Tripwire: under allow_bots=all the mention path must not be probed.
-    adapter = make_adapter_skeleton(bot_open_id="ou_self", allow_bots="all")
-    calls = 0
-
-    def _tripwire(_message):
-        nonlocal calls
-        calls += 1
-        return False
-
-    adapter._mentions_self = _tripwire
-
-    sender = make_sender(sender_type="bot", open_id="ou_peer")
-    assert adapter._admit(sender, make_message()) is None
-    assert calls == 0
-
-
-def test_admit_group_mention_checked_once_per_call():
-    # Stage 2 (mentions mode) and stage 4 (group require_mention) must not
-    # double-evaluate _mentions_self for the same admit call.
-    adapter = make_adapter_skeleton(
-        bot_open_id="ou_self", allow_bots="mentions", require_mention=True,
-        group_policy="open",
-    )
-    calls = 0
-
-    def _counting(_message):
-        nonlocal calls
-        calls += 1
-        return True
-
-    adapter._mentions_self = _counting
-
-    sender = make_sender(sender_type="bot", open_id="ou_peer")
-    assert adapter._admit(sender, make_message(chat_type="group")) is None
-    assert calls == 1
-
-
-# --- Per-group require_mention override ------------------------------------
-
-
-def test_admit_per_group_require_mention_overrides_global():
-    from gateway.platforms.feishu import FeishuGroupRule
-
-    adapter = make_adapter_skeleton(
-        bot_open_id="ou_self", require_mention=True, group_policy="open",
-    )
-    adapter._group_rules = {
-        "oc_free": FeishuGroupRule(policy="open", require_mention=False),
-    }
-    stub_mention(adapter, False)
-
-    sender = make_sender(sender_type="user", open_id="ou_human")
-    assert adapter._admit(sender, make_message(chat_id="oc_free", chat_type="group")) is None
-    assert (
-        adapter._admit(sender, make_message(chat_id="oc_other", chat_type="group"))
-        == "group_policy_rejected"
-    )
-
-
-# --- Hydration -------------------------------------------------------------
-
-
-def test_hydrate_bot_identity_populates_self_ids_from_bot_v3_info(monkeypatch):
-    import asyncio
-
-    from gateway.platforms.feishu import FeishuAdapter
-
-    adapter = object.__new__(FeishuAdapter)
-    adapter._bot_open_id = ""
-    adapter._bot_user_id = ""
-    adapter._bot_name = ""
-    adapter._allow_bots = "all"
-
-    captured = {}
-
-    def _fake_request(request):
-        captured["uri"] = getattr(request, "uri", None)
-        captured["http_method"] = getattr(request, "http_method", None)
-        return SimpleNamespace(raw=SimpleNamespace(
-            content=b'{"code":0,"bot":{"app_name":"Hermes","open_id":"ou_hydrated"}}'
-        ))
-
-    adapter._client = SimpleNamespace(request=_fake_request)
-
-    asyncio.run(adapter._hydrate_bot_identity())
-
-    assert captured["uri"] == "/open-apis/bot/v3/info"
-    assert str(captured["http_method"]).endswith("GET")
-    assert adapter._bot_open_id == "ou_hydrated"
-    assert adapter._bot_name == "Hermes"
-    # /bot/v3/info doesn't surface user_id, so _bot_user_id stays empty.
-    assert adapter._bot_user_id == ""
-
-
-def test_resolve_sender_profile_uses_open_id_for_bot_name_lookup():
-    import asyncio
-
-    from gateway.platforms.feishu import FeishuAdapter
-
-    adapter = object.__new__(FeishuAdapter)
-    adapter._client = object()
-    adapter._sender_name_cache = {}
-    seen_ids = []
-
-    async def _fake_fetch_bot_names(bot_ids):
-        seen_ids.extend(bot_ids)
-        return {"ou_peer": "Peer Bot"}
-
-    adapter._fetch_bot_names = _fake_fetch_bot_names
-
-    profile = asyncio.run(
-        adapter._resolve_sender_profile(
-            SimpleNamespace(open_id="ou_peer", user_id="u_peer", union_id="on_peer"),
-            is_bot=True,
-        )
-    )
-
-    assert seen_ids == ["ou_peer"]
-    assert profile["user_id"] == "u_peer"
-    assert profile["user_name"] == "Peer Bot"
-
-
-# --- _allow_group_message matrix -------------------------------------------
-#
-# Bot-bypass semantics: admitted bots skip allowlist/blacklist (parallel
-# human-scope filters), but channel-level locks (disabled, admin_only) and
-# admin short-circuits still apply.
-
-
-def _group_case(
-    *,
-    adapter: dict | None = None,
-    admins: set | None = None,
-    group_rules: dict | None = None,
-    sender: dict | None = None,
-    chat_id: str = "oc_1",
-    is_bot: bool = False,
-    expected: bool = False,
-):
-    return {
-        "adapter": adapter or {},
-        "admins": admins or set(),
-        "group_rules": group_rules or {},
-        "sender": sender or {},
-        "chat_id": chat_id,
-        "is_bot": is_bot,
-        "expected": expected,
-    }
-
-
-def _group_rule(policy: str, **kwargs):
-    from gateway.platforms.feishu import FeishuGroupRule
-    return FeishuGroupRule(policy=policy, **kwargs)
-
-
-_GROUP_CASES = [
-    pytest.param(
-        _group_case(
-            sender={"sender_type": "bot", "open_id": "ou_peer"},
-            is_bot=True,
-            expected=True,
-        ),
-        id="bot:bypasses_default_allowlist",
-    ),
-    pytest.param(
-        _group_case(
-            sender={"sender_type": "user", "open_id": "ou_stranger"},
-            is_bot=False,
-            expected=False,
-        ),
-        id="human:gated_by_default_allowlist",
-    ),
-    pytest.param(
-        _group_case(
-            admins={"ou_peer"},
-            sender={"sender_type": "bot", "open_id": "ou_peer"},
-            is_bot=True,
-            expected=True,
-        ),
-        id="bot:admin_short_circuit",
-    ),
-    pytest.param(
-        _group_case(
-            admins={"u_admin"},
-            sender={"sender_type": "user", "open_id": None, "user_id": "u_admin"},
-            is_bot=False,
-            expected=True,
-        ),
-        id="human:admin_via_user_id",
-    ),
-    pytest.param(
-        _group_case(
-            sender={"sender_type": "bot", "open_id": "ou_peer"},
-            is_bot=True,
-            expected=True,
-        ),
-        id="bot:allowlist_skipped",
-    ),
-    pytest.param(
-        _group_case(
-            sender={"sender_type": "app", "open_id": "ou_peer"},
-            is_bot=True,
-            expected=True,
-        ),
-        id="app:allowlist_skipped",
-    ),
-]
-
-
-# Channel-lock cases need group_rules construction; keep them in a separate
-# parametrize so we can use _group_rule() (FeishuGroupRule import).
-_GROUP_RULE_CASES = [
-    pytest.param(
-        "disabled", "bot", False,
-        id="bot:disabled_policy_blocks_even_with_bypass",
-    ),
-    pytest.param(
-        "disabled", "app", False,
-        id="app:disabled_policy_blocks_even_with_bypass",
-    ),
-    pytest.param(
-        "admin_only", "bot", False,
-        id="bot:admin_only_policy_blocks_non_admin",
-    ),
-    pytest.param(
-        "admin_only", "app", False,
-        id="app:admin_only_policy_blocks_non_admin",
-    ),
-]
-
-
-@pytest.mark.parametrize("case", _GROUP_CASES)
-def test_allow_group_message_matrix(case):
-    adapter = make_adapter_skeleton(**case["adapter"])
-    adapter._admins = case["admins"]
-    adapter._group_rules = case["group_rules"]
-    sender = make_sender(**case["sender"])
-    assert adapter._allow_group_message(
-        sender_id=sender.sender_id,
-        chat_id=case["chat_id"],
-        is_bot=case["is_bot"],
-    ) is case["expected"]
-
-
-@pytest.mark.parametrize("policy, sender_type, expected", _GROUP_RULE_CASES)
-def test_allow_group_message_channel_locks_apply_to_bots(policy, sender_type, expected):
-    adapter = make_adapter_skeleton()
-    adapter._group_rules = {"oc_locked": _group_rule(policy)}
-    sender = make_sender(sender_type=sender_type, open_id="ou_peer")
-    assert adapter._allow_group_message(
-        sender_id=sender.sender_id,
-        chat_id="oc_locked",
-        is_bot=True,
-    ) is expected
-
-
-@pytest.mark.parametrize("sender_type", ["bot", "app"])
-def test_allow_group_message_blacklist_is_human_scope_only(sender_type):
-    # blacklist is parallel to allowlist (human-scope); admitted bots bypass
-    # it. To block a specific bot, gate upstream via FEISHU_ALLOW_BOTS.
-    adapter = make_adapter_skeleton()
-    adapter._group_rules = {
-        "oc_1": _group_rule("blacklist", blacklist={"ou_peer"})
-    }
-    sender = make_sender(sender_type=sender_type, open_id="ou_peer")
-    assert adapter._allow_group_message(
-        sender_id=sender.sender_id,
-        chat_id="oc_1",
-        is_bot=True,
-    ) is True
-
-
-# --- Realistic payload smoke -----------------------------------------------
-
-
-def test_admit_accepts_realistic_bot_at_bot_group_event():
-    # Locks in the real im.message.receive_v1 payload shape under mode=mentions.
-    adapter = make_adapter_skeleton(bot_open_id="ou_self", allow_bots="mentions")
-
-    mention = SimpleNamespace(
-        key="@_user_1",
-        id=SimpleNamespace(union_id="on_mentionUnion", user_id="", open_id="ou_self"),
-        name="Hermes",
-        mentioned_type="bot",
-        tenant_key="tenant_ab",
-    )
-    message = SimpleNamespace(
-        message_id="om_realistic_bot_at_bot",
-        chat_id="oc_real",
-        chat_type="group",
-        message_type="text",
-        content='{"text":"@_user_1 hello"}',
-        mentions=[mention],
-    )
-    sender = SimpleNamespace(
-        sender_type="bot",
-        sender_id=SimpleNamespace(union_id="on_peerUnion", user_id="u_peer", open_id="ou_peer_bot"),
-        tenant_key="tenant_ab",
-    )
-
-    assert adapter._admit(sender, message) is None
-
-
-# --- Event-dispatch plumbing -----------------------------------------------
-
-
-def test_handle_message_event_data_drops_bot_sender_by_default():
-    import asyncio
-
-    adapter = make_adapter_skeleton()
-    install_dedup_state(adapter)
-    processed = []
-
-    async def _fake_process_inbound_message(**kwargs):
-        processed.append(kwargs)
-
-    adapter._process_inbound_message = _fake_process_inbound_message
-
-    data = SimpleNamespace(
-        event=SimpleNamespace(
-            sender=make_sender(sender_type="bot", open_id="ou_peer"),
-            message=make_message(message_id="om_bot_default", chat_type="p2p"),
-        )
-    )
-
-    asyncio.run(adapter._handle_message_event_data(data))
-    assert processed == []
-
-
-def test_handle_message_event_data_forwards_sender_when_admitted():
-    import asyncio
-
-    adapter = make_adapter_skeleton(allow_bots="all")
-    install_dedup_state(adapter)
-    captured = {}
-
-    async def _fake_process_inbound_message(**kwargs):
-        captured.update(kwargs)
-
-    adapter._process_inbound_message = _fake_process_inbound_message
-
-    sender = make_sender(sender_type="bot", open_id="ou_peer")
-    data = SimpleNamespace(
-        event=SimpleNamespace(
-            sender=sender,
-            message=make_message(message_id="om_bot_ok", chat_type="p2p"),
-        )
-    )
-
-    asyncio.run(adapter._handle_message_event_data(data))
-    assert captured.get("sender_id") is sender.sender_id
-    assert captured.get("is_bot") is True
-    assert captured.get("message_id") == "om_bot_ok"

tests/gateway/test_feishu_bot_auth_bypass.py

@@ -1,113 +0,0 @@
-"""Regression guard for Feishu bot-sender authorization bypass.
-
-Mirrors tests/gateway/test_discord_bot_auth_bypass.py for Platform.FEISHU.
-Without the bypass in gateway/run.py, Feishu bot senders admitted by the
-adapter would be rejected at _is_user_authorized with "Unauthorized user"
-— same class of bug as Discord #4466.
-"""
-
-from __future__ import annotations
-
-from types import SimpleNamespace
-
-import pytest
-
-from gateway.session import Platform, SessionSource
-
-
-@pytest.fixture(autouse=True)
-def _isolate_feishu_env(monkeypatch):
-    for var in (
-        "FEISHU_ALLOW_BOTS",
-        "FEISHU_ALLOWED_USERS",
-        "FEISHU_ALLOW_ALL_USERS",
-        "GATEWAY_ALLOW_ALL_USERS",
-        "GATEWAY_ALLOWED_USERS",
-    ):
-        monkeypatch.delenv(var, raising=False)
-
-
-def _make_bare_runner():
-    from gateway.run import GatewayRunner
-
-    runner = object.__new__(GatewayRunner)
-    runner.pairing_store = SimpleNamespace(is_approved=lambda *_a, **_kw: False)
-    return runner
-
-
-def _make_feishu_bot_source(open_id: str = "ou_peer"):
-    return SessionSource(
-        platform=Platform.FEISHU,
-        chat_id="oc_1",
-        chat_type="group",
-        user_id=open_id,
-        user_name="PeerBot",
-        is_bot=True,
-    )
-
-
-def _make_feishu_human_source(open_id: str = "ou_human"):
-    return SessionSource(
-        platform=Platform.FEISHU,
-        chat_id="oc_1",
-        chat_type="group",
-        user_id=open_id,
-        user_name="Human",
-        is_bot=False,
-    )
-
-
-def test_feishu_bot_authorized_when_allow_bots_mentions(monkeypatch):
-    runner = _make_bare_runner()
-    monkeypatch.setenv("FEISHU_ALLOW_BOTS", "mentions")
-    monkeypatch.setenv("FEISHU_ALLOWED_USERS", "ou_human")
-
-    assert runner._is_user_authorized(_make_feishu_bot_source("ou_peer")) is True
-
-
-def test_feishu_bot_authorized_when_allow_bots_all(monkeypatch):
-    runner = _make_bare_runner()
-    monkeypatch.setenv("FEISHU_ALLOW_BOTS", "all")
-    monkeypatch.setenv("FEISHU_ALLOWED_USERS", "ou_human")
-
-    assert runner._is_user_authorized(_make_feishu_bot_source()) is True
-
-
-def test_feishu_bot_NOT_authorized_when_allow_bots_none(monkeypatch):
-    runner = _make_bare_runner()
-    monkeypatch.setenv("FEISHU_ALLOW_BOTS", "none")
-    monkeypatch.setenv("FEISHU_ALLOWED_USERS", "ou_human")
-
-    assert runner._is_user_authorized(_make_feishu_bot_source("ou_peer")) is False
-
-
-def test_feishu_bot_NOT_authorized_when_allow_bots_unset(monkeypatch):
-    runner = _make_bare_runner()
-    monkeypatch.setenv("FEISHU_ALLOWED_USERS", "ou_human")
-
-    assert runner._is_user_authorized(_make_feishu_bot_source("ou_peer")) is False
-
-
-def test_feishu_human_still_checked_against_allowlist_when_bot_policy_set(monkeypatch):
-    """FEISHU_ALLOW_BOTS=all must NOT open the gate for humans."""
-    runner = _make_bare_runner()
-    monkeypatch.setenv("FEISHU_ALLOW_BOTS", "all")
-    monkeypatch.setenv("FEISHU_ALLOWED_USERS", "ou_human")
-
-    assert runner._is_user_authorized(_make_feishu_human_source("ou_stranger")) is False
-    assert runner._is_user_authorized(_make_feishu_human_source("ou_human")) is True
-
-
-def test_feishu_bot_bypass_does_not_leak_to_other_platforms(monkeypatch):
-    """FEISHU_ALLOW_BOTS=all must not authorize Telegram/Discord bot sources."""
-    runner = _make_bare_runner()
-    monkeypatch.setenv("FEISHU_ALLOW_BOTS", "all")
-
-    telegram_bot = SessionSource(
-        platform=Platform.TELEGRAM,
-        chat_id="123",
-        chat_type="channel",
-        user_id="999",
-        is_bot=True,
-    )
-    assert runner._is_user_authorized(telegram_bot) is False

tests/gateway/test_fresh_reset_skill_injection.py

@@ -1,201 +0,0 @@
-"""Regression tests for topic/channel skill auto-injection after /new or /reset.
-
-Covers the fix for issue #6508.
-
-Before the fix:
-    1. User sends ``/new`` — ``reset_session`` creates a fresh SessionEntry
-       with ``created_at == updated_at``.
-    2. User sends the next message.
-    3. ``get_or_create_session`` finds the entry and bumps
-       ``entry.updated_at = now`` (microseconds after ``created_at``).
-    4. ``_handle_message_with_agent`` checks
-       ``_is_new_session = (created_at == updated_at) or was_auto_reset``.
-       Both are False → ``_is_new_session = False`` → topic/channel skills
-       are silently skipped for the first message of a manually reset session.
-
-After the fix:
-    ``reset_session`` stamps the new entry with ``is_fresh_reset=True``.
-    ``_handle_message_with_agent`` ORs this into ``_is_new_session`` and
-    consumes the flag immediately after the check, so subsequent messages
-    are treated as continuing the session and the flag does not leak.
-
-We use ``was_auto_reset`` for surprise resets (idle/daily/suspended) and
-``is_fresh_reset`` for user-initiated resets because the former also drives
-a "Session automatically reset due to inactivity" user-facing notice and
-a context-note prepend into the agent's prompt — both wrong for an explicit
-/new or /reset.
-"""
-import pytest
-
-from gateway.config import GatewayConfig, Platform
-from gateway.session import SessionEntry, SessionSource, SessionStore
-
-
-def _make_store(tmp_path):
-    return SessionStore(sessions_dir=tmp_path, config=GatewayConfig())
-
-
-def _make_source(chat_id="123", user_id="u1"):
-    return SessionSource(
-        platform=Platform.TELEGRAM,
-        chat_id=chat_id,
-        user_id=user_id,
-    )
-
-
-def _is_new_session(entry) -> bool:
-    """Mirror of the predicate in ``_handle_message_with_agent``.
-
-    Kept in-sync with the production check so this test fails loudly if the
-    upstream logic regresses.
-    """
-    return (
-        entry.created_at == entry.updated_at
-        or getattr(entry, "was_auto_reset", False)
-        or getattr(entry, "is_fresh_reset", False)
-    )
-
-
-# ---------------------------------------------------------------------------
-# reset_session stamps is_fresh_reset=True
-# ---------------------------------------------------------------------------
-
-class TestResetSessionStampsFreshReset:
-    def test_reset_session_sets_is_fresh_reset_true(self, tmp_path):
-        store = _make_store(tmp_path)
-        source = _make_source()
-        store.get_or_create_session(source)
-        session_key = store._generate_session_key(source)
-
-        new_entry = store.reset_session(session_key)
-
-        assert new_entry is not None
-        assert new_entry.is_fresh_reset is True
-
-    def test_reset_session_unknown_key_returns_none(self, tmp_path):
-        store = _make_store(tmp_path)
-        assert store.reset_session("unknown:key") is None
-
-    def test_fresh_session_does_not_have_is_fresh_reset(self, tmp_path):
-        """A vanilla first-time session should not carry the flag."""
-        store = _make_store(tmp_path)
-        entry = store.get_or_create_session(_make_source())
-        assert entry.is_fresh_reset is False
-
-
-# ---------------------------------------------------------------------------
-# Core regression: _is_new_session stays True after updated_at bump
-# ---------------------------------------------------------------------------
-
-class TestIsNewSessionSurvivesUpdatedAtBump:
-    def test_is_new_session_true_after_reset_then_next_message(self, tmp_path):
-        """The actual bug: _is_new_session was False on message after /reset."""
-        store = _make_store(tmp_path)
-        source = _make_source()
-        store.get_or_create_session(source)
-        session_key = store._generate_session_key(source)
-
-        # User sends /reset
-        store.reset_session(session_key)
-
-        # Next inbound message — get_or_create_session bumps updated_at
-        entry = store.get_or_create_session(source)
-
-        # Before the fix: created_at != updated_at, was_auto_reset=False → False
-        # After the fix: is_fresh_reset=True carries the signal through the bump
-        assert _is_new_session(entry) is True
-
-    def test_flag_consumed_after_first_read(self, tmp_path):
-        """After the message handler consumes is_fresh_reset, the NEXT
-        message should not be treated as a new session (skill re-injection
-        must not fire a second time).
-        """
-        store = _make_store(tmp_path)
-        source = _make_source()
-        store.get_or_create_session(source)
-        session_key = store._generate_session_key(source)
-        store.reset_session(session_key)
-
-        # First message — handler consumes the flag
-        entry = store.get_or_create_session(source)
-        assert _is_new_session(entry) is True
-        entry.is_fresh_reset = False  # what _handle_message_with_agent does
-
-        # Second message — must not be treated as new
-        entry = store.get_or_create_session(source)
-        assert _is_new_session(entry) is False
-
-
-# ---------------------------------------------------------------------------
-# Vanilla-session behavior is unchanged
-# ---------------------------------------------------------------------------
-
-class TestVanillaBehaviorUnaffected:
-    def test_ongoing_session_not_flagged_as_new(self, tmp_path):
-        store = _make_store(tmp_path)
-        source = _make_source()
-        store.get_or_create_session(source)
-
-        # Second message on the same session — updated_at bumps,
-        # is_fresh_reset was never set
-        entry = store.get_or_create_session(source)
-        assert entry.is_fresh_reset is False
-        assert _is_new_session(entry) is False
-
-    def test_idle_auto_reset_does_not_set_is_fresh_reset(self, tmp_path):
-        """Idle/daily auto-resets use was_auto_reset — confirm they do NOT
-        also set is_fresh_reset (which would double-fire the skill path and
-        not leak through the auto-reset guard).
-        """
-        store = _make_store(tmp_path)
-        source = _make_source()
-        entry = store.get_or_create_session(source)
-
-        # Simulate the auto-reset code path: get_or_create_session's internal
-        # branch that sets was_auto_reset does NOT touch is_fresh_reset.
-        # Construct a fresh entry the same way that branch does.
-        store._entries.pop(store._generate_session_key(source))
-        fresh = SessionEntry(
-            session_key=entry.session_key,
-            session_id="new_id",
-            created_at=entry.created_at,
-            updated_at=entry.created_at,
-            origin=source,
-            was_auto_reset=True,
-            auto_reset_reason="idle",
-        )
-        assert fresh.is_fresh_reset is False
-        assert fresh.was_auto_reset is True
-
-
-# ---------------------------------------------------------------------------
-# Persistence through sessions.json round-trip
-# ---------------------------------------------------------------------------
-
-class TestPersistence:
-    def test_is_fresh_reset_survives_to_dict_from_dict(self, tmp_path):
-        """Protect against the gateway restarting between /reset and the
-        next message — the flag must be persisted in sessions.json.
-        """
-        store = _make_store(tmp_path)
-        source = _make_source()
-        store.get_or_create_session(source)
-        session_key = store._generate_session_key(source)
-        new_entry = store.reset_session(session_key)
-
-        assert new_entry.is_fresh_reset is True
-        restored = SessionEntry.from_dict(new_entry.to_dict())
-        assert restored.is_fresh_reset is True
-
-    def test_default_false_when_missing_from_dict(self, tmp_path):
-        """Older sessions.json files written before this field existed must
-        load cleanly with is_fresh_reset defaulting to False.
-        """
-        data = {
-            "session_key": "telegram:1:123",
-            "session_id": "sess1",
-            "created_at": "2026-01-01T00:00:00",
-            "updated_at": "2026-01-01T00:00:00",
-        }
-        entry = SessionEntry.from_dict(data)
-        assert entry.is_fresh_reset is False

tests/gateway/test_home_target_env_var.py

@@ -1,36 +0,0 @@
-"""Regression tests for /sethome env-var resolution.
-
-The `/sethome` command writes to a platform's home-target env var. Two platforms
-don't follow the `{PLATFORM}_HOME_CHANNEL` convention: matrix uses
-`MATRIX_HOME_ROOM` and email uses `EMAIL_HOME_ADDRESS`. Before PR #12698
-`/sethome` hardcoded the `_HOME_CHANNEL` suffix, so Matrix and Email saves went
-to env vars nothing read on startup — the home channel appeared to set
-successfully but was lost on every new gateway session.
-"""
-
-from gateway.run import _home_target_env_var
-
-
-def test_matrix_home_target_env_var_uses_home_room():
-    assert _home_target_env_var("matrix") == "MATRIX_HOME_ROOM"
-
-
-def test_email_home_target_env_var_uses_home_address():
-    assert _home_target_env_var("email") == "EMAIL_HOME_ADDRESS"
-
-
-def test_telegram_home_target_env_var_uses_home_channel():
-    assert _home_target_env_var("telegram") == "TELEGRAM_HOME_CHANNEL"
-
-
-def test_discord_home_target_env_var_uses_home_channel():
-    assert _home_target_env_var("discord") == "DISCORD_HOME_CHANNEL"
-
-
-def test_unknown_platform_home_target_env_var_falls_back_to_home_channel():
-    assert _home_target_env_var("custom") == "CUSTOM_HOME_CHANNEL"
-
-
-def test_case_insensitive_platform_name():
-    assert _home_target_env_var("MATRIX") == "MATRIX_HOME_ROOM"
-    assert _home_target_env_var("Email") == "EMAIL_HOME_ADDRESS"

tests/gateway/test_native_image_buffer_isolation.py

@@ -1,79 +0,0 @@
-import pytest
-
-from gateway.config import GatewayConfig, Platform, PlatformConfig
-from gateway.platforms.base import MessageEvent, MessageType
-from gateway.run import GatewayRunner
-from gateway.session import SessionSource, build_session_key
-
-
-def _make_runner() -> GatewayRunner:
-    runner = GatewayRunner.__new__(GatewayRunner)
-    runner.config = GatewayConfig(
-        platforms={Platform.TELEGRAM: PlatformConfig(enabled=True, token="fake")},
-    )
-    runner.adapters = {}
-    runner._model = "openai/gpt-4.1-mini"
-    runner._base_url = None
-    runner._decide_image_input_mode = lambda: "native"
-    return runner
-
-
-def _source(chat_id: str) -> SessionSource:
-    return SessionSource(
-        platform=Platform.TELEGRAM,
-        chat_id=chat_id,
-        chat_type="private",
-        user_name=f"user-{chat_id}",
-    )
-
-
-def _image_event(source: SessionSource, path: str) -> MessageEvent:
-    return MessageEvent(
-        text="see image",
-        message_type=MessageType.PHOTO,
-        source=source,
-        media_urls=[path],
-        media_types=["image/png"],
-    )
-
-
-@pytest.mark.asyncio
-async def test_native_image_buffer_isolated_per_session():
-    runner = _make_runner()
-    source_a = _source("chat-a")
-    source_b = _source("chat-b")
-
-    await runner._prepare_inbound_message_text(
-        event=_image_event(source_a, "/tmp/a.png"),
-        source=source_a,
-        history=[],
-    )
-    await runner._prepare_inbound_message_text(
-        event=_image_event(source_b, "/tmp/b.png"),
-        source=source_b,
-        history=[],
-    )
-
-    assert runner._consume_pending_native_image_paths(build_session_key(source_a)) == ["/tmp/a.png"]
-    assert runner._consume_pending_native_image_paths(build_session_key(source_b)) == ["/tmp/b.png"]
-
-
-@pytest.mark.asyncio
-async def test_native_image_buffer_not_cleared_by_other_sessions_without_images():
-    runner = _make_runner()
-    source_a = _source("chat-a")
-    source_b = _source("chat-b")
-
-    await runner._prepare_inbound_message_text(
-        event=_image_event(source_a, "/tmp/a.png"),
-        source=source_a,
-        history=[],
-    )
-    await runner._prepare_inbound_message_text(
-        event=MessageEvent(text="plain text", source=source_b),
-        source=source_b,
-        history=[],
-    )
-
-    assert runner._consume_pending_native_image_paths(build_session_key(source_a)) == ["/tmp/a.png"]
-    assert runner._consume_pending_native_image_paths(build_session_key(source_b)) == []

tests/gateway/test_reasoning_command.py

@@ -407,44 +407,3 @@ class TestReasoningCommand:
         assert result["final_response"] == "ok"
         assert _CapturingAgent.last_init is not None
         assert "homeassistant" in set(_CapturingAgent.last_init["enabled_toolsets"])
-
-
-class TestLoadShowReasoningCoercion:
-    """Regression: display.show_reasoning must be coerced, not bool()'d."""
-
-    def _load_with_config(self, tmp_path, monkeypatch, yaml_body: str) -> bool:
-        hermes_home = tmp_path / "hermes"
-        hermes_home.mkdir()
-        (hermes_home / "config.yaml").write_text(yaml_body, encoding="utf-8")
-        monkeypatch.setattr(gateway_run, "_hermes_home", hermes_home)
-        return gateway_run.GatewayRunner._load_show_reasoning()
-
-    def test_quoted_false_is_false(self, tmp_path, monkeypatch):
-        assert self._load_with_config(
-            tmp_path, monkeypatch,
-            'display:\n  show_reasoning: "false"\n',
-        ) is False
-
-    def test_quoted_off_is_false(self, tmp_path, monkeypatch):
-        assert self._load_with_config(
-            tmp_path, monkeypatch,
-            'display:\n  show_reasoning: "off"\n',
-        ) is False
-
-    def test_quoted_true_is_true(self, tmp_path, monkeypatch):
-        assert self._load_with_config(
-            tmp_path, monkeypatch,
-            'display:\n  show_reasoning: "true"\n',
-        ) is True
-
-    def test_bare_true_is_true(self, tmp_path, monkeypatch):
-        assert self._load_with_config(
-            tmp_path, monkeypatch,
-            'display:\n  show_reasoning: true\n',
-        ) is True
-
-    def test_missing_is_false(self, tmp_path, monkeypatch):
-        assert self._load_with_config(
-            tmp_path, monkeypatch,
-            'display: {}\n',
-        ) is False

tests/gateway/test_restart_notification.py

@@ -113,36 +113,6 @@ async def test_restart_command_preserves_thread_id(tmp_path, monkeypatch):
     assert data["thread_id"] == "topic_7"
 
 
-@pytest.mark.asyncio
-async def test_restart_command_uses_atomic_json_writes_for_marker_files(tmp_path, monkeypatch):
-    monkeypatch.setattr(gateway_run, "_hermes_home", tmp_path)
-
-    calls = []
-
-    def _fake_atomic_json_write(path, payload, **kwargs):
-        calls.append((Path(path).name, payload, kwargs))
-
-    monkeypatch.setattr(gateway_run, "atomic_json_write", _fake_atomic_json_write)
-
-    runner, _adapter = make_restart_runner()
-    runner.request_restart = MagicMock(return_value=True)
-
-    source = make_restart_source(chat_id="42")
-    event = MessageEvent(
-        text="/restart",
-        message_type=MessageType.TEXT,
-        source=source,
-        message_id="m1",
-    )
-
-    await runner._handle_restart_command(event)
-
-    names = [name for name, _payload, _kwargs in calls]
-    assert names == [".restart_notify.json", ".restart_last_processed.json"]
-    assert calls[0][1]["chat_id"] == "42"
-    assert calls[1][1]["platform"] == "telegram"
-
-
 # ── _send_restart_notification ───────────────────────────────────────────
 
 

tests/gateway/test_restart_resume_pending.py

@@ -999,65 +999,3 @@ class TestStuckLoopEscalation:
 
         assert store._entries[entry.session_key].resume_pending is False
         assert not counts_file.exists()
-
-    def test_increment_restart_failure_counts_uses_atomic_json_write(
-        self, tmp_path, monkeypatch
-    ):
-        from gateway.run import GatewayRunner
-
-        source = _make_source()
-        session_key = _make_store(tmp_path).get_or_create_session(source).session_key
-
-        monkeypatch.setattr("gateway.run._hermes_home", tmp_path)
-        calls = []
-
-        def _fake_atomic_json_write(path, payload, **kwargs):
-            calls.append((path, payload, kwargs))
-
-        monkeypatch.setattr("gateway.run.atomic_json_write", _fake_atomic_json_write)
-
-        runner = object.__new__(GatewayRunner)
-        runner._increment_restart_failure_counts({session_key})
-
-        assert calls == [
-            (
-                tmp_path / ".restart_failure_counts",
-                {session_key: 1},
-                {"indent": None},
-            )
-        ]
-
-    def test_clear_restart_failure_count_uses_atomic_json_write_when_entries_remain(
-        self, tmp_path, monkeypatch
-    ):
-        import json
-
-        from gateway.run import GatewayRunner
-
-        source = _make_source()
-        session_key = _make_store(tmp_path).get_or_create_session(source).session_key
-        other_key = "agent:main:telegram:dm:other"
-        counts_file = tmp_path / ".restart_failure_counts"
-        counts_file.write_text(
-            json.dumps({session_key: 2, other_key: 1}),
-            encoding="utf-8",
-        )
-
-        monkeypatch.setattr("gateway.run._hermes_home", tmp_path)
-        calls = []
-
-        def _fake_atomic_json_write(path, payload, **kwargs):
-            calls.append((path, payload, kwargs))
-
-        monkeypatch.setattr("gateway.run.atomic_json_write", _fake_atomic_json_write)
-
-        runner = object.__new__(GatewayRunner)
-        runner._clear_restart_failure_count(session_key)
-
-        assert calls == [
-            (
-                tmp_path / ".restart_failure_counts",
-                {other_key: 1},
-                {"indent": None},
-            )
-        ]

tests/gateway/test_session.py

@@ -1243,7 +1243,7 @@ class TestRewriteTranscriptPreservesReasoning:
         assert after[0].get("reasoning_details") == [{"type": "summary", "text": "step by step"}]
         assert after[0].get("codex_reasoning_items") == [{"id": "r1", "type": "reasoning"}]
 
-    def test_db_rewrite_is_atomic_on_insert_failure(self, tmp_path, monkeypatch):
+    def test_db_rewrite_is_atomic_on_insert_failure(self, tmp_path):
         from hermes_state import SessionDB
 
         db = SessionDB(db_path=tmp_path / "test.db")
@@ -1258,27 +1258,16 @@ class TestRewriteTranscriptPreservesReasoning:
         store._db = db
         store._loaded = True
 
-        # Force the second insert inside replace_messages to fail, simulating
-        # any storage-layer error that might abort a multi-row rewrite.
-        real_encode = SessionDB._encode_content
-        calls = {"n": 0}
-
-        def flaky_encode(cls, content):
-            calls["n"] += 1
-            if calls["n"] == 2:
-                raise RuntimeError("simulated storage failure")
-            return real_encode.__func__(cls, content)
-
-        monkeypatch.setattr(SessionDB, "_encode_content", classmethod(flaky_encode))
-
         replacement = [
             {"role": "user", "content": "after user"},
-            {"role": "assistant", "content": "after assistant"},
+            {
+                "role": "assistant",
+                "content": {"not": "sqlite-bindable but JSONL-safe"},
+            },
         ]
 
         store.rewrite_transcript(session_id, replacement)
 
-        # The rewrite must roll back atomically — original messages preserved.
         after = db.get_messages_as_conversation(session_id)
         assert [msg["content"] for msg in after] == [
             "before user",

tests/gateway/test_session_boundary_security_state.py

@@ -10,7 +10,6 @@ from gateway.platforms.base import MessageEvent
 from gateway.session import SessionEntry, SessionSource, build_session_key
 from tools import approval as approval_mod
 from tools.approval import (
-    _ApprovalEntry,
     approve_session,
     enable_session_yolo,
     is_approved,
@@ -173,38 +172,6 @@ async def test_branch_clears_session_scoped_approval_and_yolo_state():
     assert other_key in runner._update_prompt_pending
 
 
-@pytest.mark.asyncio
-async def test_branch_preserves_persisted_assistant_metadata():
-    runner, _session_key = _make_branch_runner()
-    runner.session_store.load_transcript.return_value = [
-        {"role": "user", "content": "hello"},
-        {
-            "role": "assistant",
-            "content": "world",
-            "finish_reason": "stop",
-            "reasoning": "thinking",
-            "reasoning_content": "provider scratchpad",
-            "reasoning_details": [{"type": "summary", "text": "step"}],
-            "codex_reasoning_items": [{"id": "r1", "type": "reasoning"}],
-            "codex_message_items": [{"id": "m1", "type": "message"}],
-        },
-    ]
-
-    result = await runner._handle_branch_command(_make_event("/branch"))
-
-    assert "Branched to" in result
-    append_calls = runner._session_db.append_message.call_args_list
-    assert len(append_calls) == 2
-    assistant_kwargs = append_calls[1].kwargs
-    assert assistant_kwargs["role"] == "assistant"
-    assert assistant_kwargs["finish_reason"] == "stop"
-    assert assistant_kwargs["reasoning"] == "thinking"
-    assert assistant_kwargs["reasoning_content"] == "provider scratchpad"
-    assert assistant_kwargs["reasoning_details"] == [{"type": "summary", "text": "step"}]
-    assert assistant_kwargs["codex_reasoning_items"] == [{"id": "r1", "type": "reasoning"}]
-    assert assistant_kwargs["codex_message_items"] == [{"id": "m1", "type": "message"}]
-
-
 def test_clear_session_boundary_security_state_is_scoped():
     """The helper must wipe only the target session's approval/yolo state.
 
@@ -247,30 +214,3 @@ def test_clear_session_boundary_security_state_is_scoped():
     runner._clear_session_boundary_security_state("")
     assert is_approved(other_key, "recursive delete") is True
     assert other_key in runner._update_prompt_pending
-
-
-def test_clear_session_boundary_security_state_wakes_blocked_approvals():
-    """Boundary cleanup must cancel blocked approval waiters immediately."""
-    from gateway.run import GatewayRunner
-
-    runner = object.__new__(GatewayRunner)
-    runner._pending_approvals = {}
-    runner._update_prompt_pending = {}
-
-    source = _make_source()
-    session_key = build_session_key(source)
-    other_key = "agent:main:telegram:dm:other-chat"
-
-    target_entry = _ApprovalEntry({"command": "rm -rf /tmp/demo"})
-    other_entry = _ApprovalEntry({"command": "rm -rf /tmp/other"})
-    approval_mod._gateway_queues[session_key] = [target_entry]
-    approval_mod._gateway_queues[other_key] = [other_entry]
-
-    runner._clear_session_boundary_security_state(session_key)
-
-    assert target_entry.event.is_set()
-    assert target_entry.result == "deny"
-    assert other_entry.event.is_set() is False
-    assert other_entry.result is None
-    assert session_key not in approval_mod._gateway_queues
-    assert other_key in approval_mod._gateway_queues

tests/gateway/test_session_race_guard.py

@@ -226,39 +226,6 @@ def test_merge_pending_message_event_merges_text_and_photo_followups():
     assert merged.media_types == ["image/png"]
 
 
-def test_merge_pending_message_event_promotes_document_followups_over_text():
-    pending = {}
-    source = SessionSource(
-        platform=Platform.TELEGRAM,
-        chat_id="12345",
-        chat_type="dm",
-        user_id="u1",
-    )
-    session_key = build_session_key(source)
-
-    text_event = MessageEvent(
-        text="please review this",
-        message_type=MessageType.TEXT,
-        source=source,
-    )
-    document_event = MessageEvent(
-        text="",
-        message_type=MessageType.DOCUMENT,
-        source=source,
-        media_urls=["/tmp/report.pdf"],
-        media_types=["application/pdf"],
-    )
-
-    merge_pending_message_event(pending, session_key, text_event, merge_text=True)
-    merge_pending_message_event(pending, session_key, document_event, merge_text=True)
-
-    merged = pending[session_key]
-    assert merged.message_type == MessageType.DOCUMENT
-    assert merged.text == "please review this"
-    assert merged.media_urls == ["/tmp/report.pdf"]
-    assert merged.media_types == ["application/pdf"]
-
-
 @pytest.mark.asyncio
 async def test_recent_telegram_text_followup_is_queued_without_interrupt():
     runner = _make_runner()

tests/gateway/test_signal.py

@@ -1649,148 +1649,3 @@ class TestSignalSendTimeout:
         # 32 attachments × 5s = 160s; ought to comfortably outlast a
         # serial upload of an attachment-heavy batch.
         assert _signal_send_timeout(32) == 160.0
-
-
-# ---------------------------------------------------------------------------
-# Contentless Envelope Filtering (profile key updates, empty messages)
-# ---------------------------------------------------------------------------
-
-class TestSignalContentlessEnvelope:
-    """Verify that profile key updates and empty Signal messages are skipped."""
-
-    @pytest.mark.asyncio
-    async def test_skips_profile_key_update_no_message_field(self, monkeypatch):
-        """Profile key updates may carry a dataMessage without 'message' field.
-        Must be skipped to avoid triggering agent turns for metadata."""
-        adapter = _make_signal_adapter(monkeypatch)
-        captured = {}
-
-        async def fake_handle(event):
-            captured["event"] = event
-
-        adapter.handle_message = fake_handle
-
-        # Profile key update: dataMessage exists but has no "message" field
-        await adapter._handle_envelope({
-            "envelope": {
-                "sourceNumber": "+155****9999",
-                "sourceUuid": "05668cf3-8ffa-467e-9b24-f5eefa5cf475",
-                "sourceName": "Elliott McManis",
-                "timestamp": 1777600696077,
-                "dataMessage": {
-                    # No "message" field — profile key update metadata only
-                    "profileKey": "some-profile-key-data",
-                },
-            }
-        })
-
-        assert "event" not in captured, "Profile key update should be skipped"
-
-    @pytest.mark.asyncio
-    async def test_skips_empty_message(self, monkeypatch):
-        """Empty text messages (message='') should be skipped."""
-        adapter = _make_signal_adapter(monkeypatch)
-        captured = {}
-
-        async def fake_handle(event):
-            captured["event"] = event
-
-        adapter.handle_message = fake_handle
-
-        await adapter._handle_envelope({
-            "envelope": {
-                "sourceNumber": "+155****9999",
-                "sourceUuid": "05668cf3-8ffa-467e-9b24-f5eefa5cf475",
-                "sourceName": "Elliott McManis",
-                "timestamp": 1777600696077,
-                "dataMessage": {
-                    "message": "",
-                },
-            }
-        })
-
-        assert "event" not in captured, "Empty message should be skipped"
-
-    @pytest.mark.asyncio
-    async def test_skips_whitespace_only_message(self, monkeypatch):
-        """Whitespace-only messages ('   ') should be skipped."""
-        adapter = _make_signal_adapter(monkeypatch)
-        captured = {}
-
-        async def fake_handle(event):
-            captured["event"] = event
-
-        adapter.handle_message = fake_handle
-
-        await adapter._handle_envelope({
-            "envelope": {
-                "sourceNumber": "+155****9999",
-                "sourceUuid": "05668cf3-8ffa-467e-9b24-f5eefa5cf475",
-                "sourceName": "Elliott McManis",
-                "timestamp": 1777600696077,
-                "dataMessage": {
-                    "message": "   \n\t  ",
-                },
-            }
-        })
-
-        assert "event" not in captured, "Whitespace-only message should be skipped"
-
-    @pytest.mark.asyncio
-    async def test_allows_message_with_attachment_no_text(self, monkeypatch):
-        """Messages with attachments but no text should still be processed."""
-        adapter = _make_signal_adapter(monkeypatch)
-        captured = {}
-
-        async def fake_handle(event):
-            captured["event"] = event
-
-        adapter.handle_message = fake_handle
-
-        # Mock attachment fetch to return a cached image
-        png_data = b"\x89PNG\r\n\x1a\n" + b"\x00" * 100
-        b64_data = base64.b64encode(png_data).decode()
-        adapter._rpc, _ = _stub_rpc({"data": b64_data})
-
-        with patch("gateway.platforms.signal.cache_image_from_bytes", return_value="/tmp/img.png"):
-            await adapter._handle_envelope({
-                "envelope": {
-                    "sourceNumber": "+155****9999",
-                    "sourceUuid": "05668cf3-8ffa-467e-9b24-f5eefa5cf475",
-                    "sourceName": "Elliott McManis",
-                    "timestamp": 1777600696077,
-                    "dataMessage": {
-                        "message": "",  # No text
-                        "attachments": [{"id": "att-123", "size": 200}],
-                    },
-                }
-            })
-
-        assert "event" in captured, "Message with attachment should NOT be skipped"
-        assert captured["event"].media_urls == ["/tmp/img.png"]
-
-    @pytest.mark.asyncio
-    async def test_allows_normal_text_message(self, monkeypatch):
-        """Normal text messages should still flow through."""
-        adapter = _make_signal_adapter(monkeypatch)
-        captured = {}
-
-        async def fake_handle(event):
-            captured["event"] = event
-
-        adapter.handle_message = fake_handle
-
-        await adapter._handle_envelope({
-            "envelope": {
-                "sourceNumber": "+155****9999",
-                "sourceUuid": "05668cf3-8ffa-467e-9b24-f5eefa5cf475",
-                "sourceName": "Elliott McManis",
-                "timestamp": 1777600696077,
-                "dataMessage": {
-                    "message": "hello world",
-                },
-            }
-        })
-
-        assert "event" in captured, "Normal message should NOT be skipped"
-        assert captured["event"].text == "hello world"

tests/gateway/test_status.py

@@ -2,7 +2,6 @@
 
 import json
 import os
-from pathlib import Path
 from types import SimpleNamespace
 
 from gateway import status
@@ -246,27 +245,6 @@ class TestGatewayPidState:
 
 
 class TestGatewayRuntimeStatus:
-    def test_write_json_file_uses_atomic_json_write(self, tmp_path, monkeypatch):
-        monkeypatch.setenv("HERMES_HOME", str(tmp_path))
-        calls = []
-
-        def _fake_atomic_json_write(path, payload, **kwargs):
-            calls.append((Path(path), payload, kwargs))
-
-        monkeypatch.setattr(status, "atomic_json_write", _fake_atomic_json_write)
-
-        payload = {"gateway_state": "running"}
-        target = tmp_path / "gateway_state.json"
-        status._write_json_file(target, payload)
-
-        assert calls == [
-            (
-                target,
-                payload,
-                {"indent": None, "separators": (",", ":")},
-            )
-        ]
-
     def test_write_runtime_status_overwrites_stale_pid_on_restart(self, tmp_path, monkeypatch):
         """Regression: setdefault() preserved stale PID from previous process (#1631)."""
         monkeypatch.setenv("HERMES_HOME", str(tmp_path))
@@ -371,35 +349,6 @@ class TestTerminatePid:
 
 
 class TestScopedLocks:
-    def test_windows_file_lock_uses_high_offset(self, tmp_path, monkeypatch):
-        lock_path = tmp_path / "gateway.lock"
-        handle = open(lock_path, "a+", encoding="utf-8")
-        fd = handle.fileno()
-        calls = []
-
-        def fake_locking(fd, mode, size):
-            calls.append((fd, mode, size, handle.tell()))
-
-        monkeypatch.setattr(status, "_IS_WINDOWS", True)
-        monkeypatch.setattr(
-            status,
-            "msvcrt",
-            SimpleNamespace(LK_NBLCK=1, LK_UNLCK=2, locking=fake_locking),
-            raising=False,
-        )
-
-        try:
-            assert status._try_acquire_file_lock(handle) is True
-            status._release_file_lock(handle)
-        finally:
-            handle.close()
-
-        assert calls == [
-            (fd, 1, 1, status._WINDOWS_LOCK_OFFSET),
-            (fd, 2, 1, status._WINDOWS_LOCK_OFFSET),
-        ]
-        assert lock_path.read_text(encoding="utf-8") == "\n"
-
     def test_acquire_scoped_lock_rejects_live_other_process(self, tmp_path, monkeypatch):
         monkeypatch.setenv("HERMES_GATEWAY_LOCK_DIR", str(tmp_path / "locks"))
         lock_path = tmp_path / "locks" / "telegram-bot-token-2bb80d537b1da3e3.lock"

tests/gateway/test_status_command.py

@@ -55,9 +55,6 @@ def _make_runner(session_entry: SessionEntry, *, platform: Platform = Platform.T
     runner._pending_approvals = {}
     runner._session_db = MagicMock()
     runner._session_db.get_session_title.return_value = None
-    # Default: no DB row → /status reports 0 tokens.  Tests that exercise
-    # the populated path override this.
-    runner._session_db.get_session.return_value = None
     runner._reasoning_config = None
     runner._provider_routing = {}
     runner._fallback_model = None
@@ -83,14 +80,6 @@ async def test_status_command_reports_running_agent_without_interrupt(monkeypatc
         total_tokens=321,
     )
     runner = _make_runner(session_entry)
-    # Token total comes from the SQLite SessionDB, not SessionEntry.
-    runner._session_db.get_session.return_value = {
-        "input_tokens": 200,
-        "output_tokens": 121,
-        "cache_read_tokens": 0,
-        "cache_write_tokens": 0,
-        "reasoning_tokens": 0,
-    }
     running_agent = MagicMock()
     runner._running_agents[build_session_key(_make_source())] = running_agent
 
@@ -124,56 +113,6 @@ async def test_status_command_includes_session_title_when_present():
     assert "**Title:** My titled session" in result
 
 
-@pytest.mark.asyncio
-async def test_status_command_reads_token_totals_from_session_db():
-    """Regression test for #17158: /status must source token totals from the
-    SQLite SessionDB (where run_agent.py persists them) and sum all component
-    counts, not from SessionEntry (which the agent never writes)."""
-    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",
-        total_tokens=0,  # SessionEntry never gets written to — always 0.
-    )
-    runner = _make_runner(session_entry)
-    runner._session_db.get_session.return_value = {
-        "input_tokens": 1000,
-        "output_tokens": 250,
-        "cache_read_tokens": 500,
-        "cache_write_tokens": 100,
-        "reasoning_tokens": 50,
-    }
-
-    result = await runner._handle_message(_make_event("/status"))
-
-    # 1000 + 250 + 500 + 100 + 50 = 1,900
-    assert "**Tokens:** 1,900" in result
-
-
-@pytest.mark.asyncio
-async def test_status_command_tokens_zero_when_session_db_row_missing():
-    """When the SessionDB has no row for the current session yet (fresh
-    session, no agent calls), /status reports 0 without raising."""
-    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",
-        total_tokens=999,  # This should be ignored.
-    )
-    runner = _make_runner(session_entry)
-    runner._session_db.get_session.return_value = None
-
-    result = await runner._handle_message(_make_event("/status"))
-
-    assert "**Tokens:** 0" in result
-
-
 @pytest.mark.asyncio
 async def test_agents_command_reports_active_agents_and_processes(monkeypatch):
     session_key = build_session_key(_make_source())
@@ -568,68 +507,3 @@ async def test_profile_command_reports_custom_root_profile(monkeypatch, tmp_path
 
     assert "**Profile:** `coder`" in result
     assert f"**Home:** `{profile_home}`" in result
-
-
-@pytest.mark.asyncio
-async def test_post_delivery_callback_generation_snapshot_happens_after_bind():
-    """Regression: the callback_generation snapshot in _process_message_background
-    must happen AFTER the handler runs, not before.
-
-    _hermes_run_generation is set on the interrupt event by
-    GatewayRunner._bind_adapter_run_generation during _handle_message_with_agent.
-    The earlier snapshot-at-task-start always captured None, which bypassed the
-    generation-ownership check in pop_post_delivery_callback and let stale runs
-    fire a fresher run's callbacks.
-    """
-    import asyncio
-    from gateway.platforms.base import BasePlatformAdapter
-
-    source = _make_source()
-    session_key = build_session_key(source)
-    fired = []
-
-    class _ConcreteAdapter(BasePlatformAdapter):
-        platform = Platform.TELEGRAM
-
-        async def connect(self): pass
-        async def disconnect(self): pass
-        async def send(self, chat_id, content, **kwargs): pass
-        async def get_chat_info(self, chat_id): return {}
-
-    adapter = _ConcreteAdapter(
-        PlatformConfig(enabled=True, token="***"), Platform.TELEGRAM
-    )
-
-    async def fake_handler(event):
-        # Simulate what _bind_adapter_run_generation does mid-run.
-        interrupt_event = adapter._active_sessions.get(session_key)
-        setattr(interrupt_event, "_hermes_run_generation", 1)
-        # Stale run registers its callback at generation=1.
-        adapter.register_post_delivery_callback(
-            session_key,
-            lambda: fired.append("older"),
-            generation=1,
-        )
-        # A fresher run overwrites with generation=2 (different dict entry).
-        adapter.register_post_delivery_callback(
-            session_key,
-            lambda: fired.append("newer"),
-            generation=2,
-        )
-        return None
-
-    adapter.set_message_handler(fake_handler)
-    event = MessageEvent(text="hello", source=source, message_id="m1")
-
-    await adapter.handle_message(event)
-    tasks = list(adapter._background_tasks)
-    assert tasks, "expected background task to be created"
-    await asyncio.gather(*tasks)
-
-    # The stale run (generation=1) must NOT fire the fresher run's callback
-    # (generation=2). With the pre-fix code, callback_generation was snapshotted
-    # as None before the handler ran, bypassing the ownership check and firing
-    # "newer" anyway.
-    assert fired == []
-    assert session_key in adapter._post_delivery_callbacks
-    assert adapter._post_delivery_callbacks[session_key][0] == 2

tests/gateway/test_telegram_approval_buttons.py

@@ -59,21 +59,6 @@ def _make_adapter(extra=None):
     return adapter
 
 
-class _AuthRunner:
-    """Minimal runner shim for callback auth tests."""
-
-    def __init__(self, authorized: bool):
-        self.authorized = authorized
-        self.last_source = None
-
-    async def _handle_message(self, event):
-        return None
-
-    def _is_user_authorized(self, source):
-        self.last_source = source
-        return self.authorized
-
-
 # ===========================================================================
 # send_exec_approval — inline keyboard buttons
 # ===========================================================================
@@ -245,41 +230,6 @@ class TestTelegramApprovalCallback:
         edit_kwargs = query.edit_message_text.call_args[1]
         assert "Denied" in edit_kwargs["text"]
 
-    @pytest.mark.asyncio
-    async def test_approval_callback_rejects_user_blocked_by_global_allowlist(self):
-        adapter = _make_adapter()
-        adapter._approval_state[7] = "agent:main:telegram:group:12345:99"
-        runner = _AuthRunner(authorized=False)
-        adapter._message_handler = runner._handle_message
-
-        query = AsyncMock()
-        query.data = "ea:once:7"
-        query.message = MagicMock()
-        query.message.chat_id = 12345
-        query.message.chat.type = "private"
-        query.from_user = MagicMock()
-        query.from_user.id = 222
-        query.from_user.first_name = "Mallory"
-        query.answer = AsyncMock()
-        query.edit_message_text = AsyncMock()
-
-        update = MagicMock()
-        update.callback_query = query
-        context = MagicMock()
-
-        with patch("tools.approval.resolve_gateway_approval") as mock_resolve:
-            await adapter._handle_callback_query(update, context)
-
-        mock_resolve.assert_not_called()
-        query.answer.assert_called_once()
-        assert "not authorized" in query.answer.call_args[1]["text"].lower()
-        query.edit_message_text.assert_not_called()
-        assert adapter._approval_state[7] == "agent:main:telegram:group:12345:99"
-        assert runner.last_source is not None
-        assert runner.last_source.platform == Platform.TELEGRAM
-        assert runner.last_source.user_id == "222"
-        assert runner.last_source.chat_id == "12345"
-
     @pytest.mark.asyncio
     async def test_already_resolved(self):
         adapter = _make_adapter()
@@ -383,39 +333,6 @@ class TestTelegramApprovalCallback:
         query.edit_message_text.assert_not_called()
         assert not (tmp_path / ".update_response").exists()
 
-    @pytest.mark.asyncio
-    async def test_update_prompt_callback_rejects_user_blocked_by_global_allowlist(self, tmp_path):
-        adapter = _make_adapter()
-        runner = _AuthRunner(authorized=False)
-        adapter._message_handler = runner._handle_message
-
-        query = AsyncMock()
-        query.data = "update_prompt:y"
-        query.message = MagicMock()
-        query.message.chat_id = 12345
-        query.message.chat.type = "private"
-        query.from_user = MagicMock()
-        query.from_user.id = 222
-        query.from_user.first_name = "Mallory"
-        query.answer = AsyncMock()
-        query.edit_message_text = AsyncMock()
-
-        update = MagicMock()
-        update.callback_query = query
-        context = MagicMock()
-
-        with patch("hermes_constants.get_hermes_home", return_value=tmp_path):
-            with patch.dict(os.environ, {"TELEGRAM_ALLOWED_USERS": ""}):
-                await adapter._handle_callback_query(update, context)
-
-        query.answer.assert_called_once()
-        assert "not authorized" in query.answer.call_args[1]["text"].lower()
-        query.edit_message_text.assert_not_called()
-        assert not (tmp_path / ".update_response").exists()
-        assert runner.last_source is not None
-        assert runner.last_source.platform == Platform.TELEGRAM
-        assert runner.last_source.user_id == "222"
-
     @pytest.mark.asyncio
     async def test_update_prompt_callback_allows_authorized_user(self, tmp_path):
         """Allowed Telegram users can still answer update prompt buttons."""

tests/gateway/test_update_command.py

@@ -17,14 +17,13 @@ from gateway.session import SessionSource
 
 
 def _make_event(text="/update", platform=Platform.TELEGRAM,
-                user_id="12345", chat_id="67890", thread_id=None):
+                user_id="12345", chat_id="67890"):
     """Build a MessageEvent for testing."""
     source = SessionSource(
         platform=platform,
         user_id=user_id,
         chat_id=chat_id,
         user_name="testuser",
-        thread_id=thread_id,
     )
     return MessageEvent(text=text, source=source)
 
@@ -215,34 +214,6 @@ class TestHandleUpdateCommand:
         assert "timestamp" in data
         assert not (hermes_home / ".update_exit_code").exists()
 
-    @pytest.mark.asyncio
-    async def test_writes_pending_marker_with_thread_id(self, tmp_path):
-        """Persists thread_id so update notifications can route back to the thread."""
-        runner = _make_runner()
-        event = _make_event(
-            platform=Platform.TELEGRAM,
-            chat_id="99999",
-            thread_id="777",
-        )
-
-        fake_root = tmp_path / "project"
-        fake_root.mkdir()
-        (fake_root / ".git").mkdir()
-        (fake_root / "gateway").mkdir()
-        (fake_root / "gateway" / "run.py").touch()
-        fake_file = str(fake_root / "gateway" / "run.py")
-        hermes_home = tmp_path / "hermes"
-        hermes_home.mkdir()
-
-        with patch("gateway.run._hermes_home", hermes_home), \
-             patch("gateway.run.__file__", fake_file), \
-             patch("shutil.which", side_effect=lambda x: "/usr/bin/hermes" if x == "hermes" else "/usr/bin/setsid"), \
-             patch("subprocess.Popen"):
-            await runner._handle_update_command(event)
-
-        data = json.loads((hermes_home / ".update_pending.json").read_text())
-        assert data["thread_id"] == "777"
-
     @pytest.mark.asyncio
     async def test_spawns_setsid(self, tmp_path):
         """Uses setsid when available."""
@@ -461,31 +432,6 @@ class TestSendUpdateNotification:
         assert call_args[0][0] == "67890"  # chat_id
         assert "Update complete" in call_args[0][1] or "update finished" in call_args[0][1].lower()
 
-    @pytest.mark.asyncio
-    async def test_sends_notification_with_thread_metadata(self, tmp_path):
-        """Final update notification preserves thread metadata when present."""
-        runner = _make_runner()
-        hermes_home = tmp_path / "hermes"
-        hermes_home.mkdir()
-
-        pending = {
-            "platform": "telegram",
-            "chat_id": "67890",
-            "thread_id": "777",
-            "user_id": "12345",
-        }
-        (hermes_home / ".update_pending.json").write_text(json.dumps(pending))
-        (hermes_home / ".update_output.txt").write_text("done")
-        (hermes_home / ".update_exit_code").write_text("0")
-
-        mock_adapter = AsyncMock()
-        runner.adapters = {Platform.TELEGRAM: mock_adapter}
-
-        with patch("gateway.run._hermes_home", hermes_home):
-            await runner._send_update_notification()
-
-        assert mock_adapter.send.call_args.kwargs["metadata"] == {"thread_id": "777"}
-
     @pytest.mark.asyncio
     async def test_strips_ansi_codes(self, tmp_path):
         """ANSI escape codes are removed from output."""

tests/gateway/test_update_streaming.py

@@ -321,58 +321,6 @@ class TestWatchUpdateProgress:
         # Check session was marked as having pending prompt
         # (may be cleared by the time we check since update finished)
 
-    @pytest.mark.asyncio
-    async def test_prompt_forwarding_preserves_thread_metadata(self, tmp_path):
-        """Forwarded update prompts keep the originating thread/topic metadata."""
-        runner = _make_runner()
-        hermes_home = tmp_path / "hermes"
-        hermes_home.mkdir()
-
-        pending = {
-            "platform": "telegram",
-            "chat_id": "111",
-            "thread_id": "777",
-            "user_id": "222",
-            "session_key": "agent:main:telegram:group:111:777",
-        }
-        (hermes_home / ".update_pending.json").write_text(json.dumps(pending))
-        (hermes_home / ".update_output.txt").write_text("")
-        (hermes_home / ".update_prompt.json").write_text(json.dumps({
-            "prompt": "Restore local changes? [Y/n]",
-            "default": "y",
-            "id": "threaded-prompt",
-        }))
-
-        class _PromptCapableAdapter:
-            def __init__(self):
-                self.send = AsyncMock()
-                self.prompt_calls = AsyncMock()
-
-            async def send_update_prompt(self, **kwargs):
-                return await self.prompt_calls(**kwargs)
-
-        mock_adapter = _PromptCapableAdapter()
-        runner.adapters = {Platform.TELEGRAM: mock_adapter}
-
-        async def finish_after_prompt():
-            await asyncio.sleep(0.3)
-            (hermes_home / ".update_response").write_text("y")
-            await asyncio.sleep(0.2)
-            (hermes_home / ".update_exit_code").write_text("0")
-
-        with patch("gateway.run._hermes_home", hermes_home):
-            task = asyncio.create_task(finish_after_prompt())
-            await runner._watch_update_progress(
-                poll_interval=0.1,
-                stream_interval=0.2,
-                timeout=5.0,
-            )
-            await task
-
-        assert mock_adapter.prompt_calls.call_args.kwargs["metadata"] == {
-            "thread_id": "777"
-        }
-
     @pytest.mark.asyncio
     async def test_cleans_up_on_completion(self, tmp_path):
         """All marker files are cleaned up when update finishes."""

tests/gateway/test_verbose_command.py

@@ -85,25 +85,6 @@ class TestVerboseCommand:
         saved = yaml.safe_load(config_path.read_text(encoding="utf-8"))
         assert saved["display"]["platforms"]["telegram"]["tool_progress"] == "verbose"
 
-    @pytest.mark.asyncio
-    async def test_quoted_false_keeps_command_disabled(self, tmp_path, monkeypatch):
-        """Quoted false must not enable the /verbose gateway command."""
-        hermes_home = tmp_path / "hermes"
-        hermes_home.mkdir()
-        config_path = hermes_home / "config.yaml"
-        config_path.write_text(
-            'display:\n  tool_progress_command: "false"\n  tool_progress: all\n',
-            encoding="utf-8",
-        )
-
-        monkeypatch.setattr(gateway_run, "_hermes_home", hermes_home)
-
-        runner = _make_runner()
-        result = await runner._handle_verbose_command(_make_event())
-
-        assert "not enabled" in result.lower()
-        assert "tool_progress_command" in result
-
     @pytest.mark.asyncio
     async def test_cycles_through_all_modes(self, tmp_path, monkeypatch):
         """Calling /verbose repeatedly cycles through all four modes."""

tests/hermes_cli/test_auth_commands.py

@@ -5,10 +5,8 @@ from __future__ import annotations
 import base64
 import json
 from datetime import datetime, timezone
-from unittest.mock import patch
 
 import pytest
-import yaml
 
 
 def _write_auth_store(tmp_path, payload: dict) -> None:
@@ -591,39 +589,6 @@ def test_logout_clears_stale_active_codex_without_provider_credentials(tmp_path,
     assert "provider: auto" in config_text
 
 
-def test_reset_config_provider_uses_atomic_yaml_write(tmp_path, monkeypatch):
-    """Logout config reset should delegate the YAML write atomically."""
-    hermes_home = tmp_path / "hermes"
-    hermes_home.mkdir(parents=True, exist_ok=True)
-    monkeypatch.setenv("HERMES_HOME", str(hermes_home))
-    config_path = hermes_home / "config.yaml"
-    original = {
-        "model": {
-            "default": "gpt-5.3-codex",
-            "provider": "openai-codex",
-            "base_url": "https://chatgpt.com/backend-api/codex",
-        }
-    }
-    config_path.write_text(yaml.safe_dump(original, sort_keys=False), encoding="utf-8")
-    original_text = config_path.read_text(encoding="utf-8")
-
-    from hermes_cli.auth import _reset_config_provider
-
-    def _boom(path, data, **kwargs):
-        assert path == config_path
-        assert data["model"]["provider"] == "auto"
-        assert data["model"]["base_url"] == "https://openrouter.ai/api/v1"
-        assert kwargs["sort_keys"] is False
-        raise OSError("simulated atomic write failure")
-
-    with patch("hermes_cli.auth.atomic_yaml_write", side_effect=_boom) as mock_write:
-        with pytest.raises(OSError, match="simulated atomic write failure"):
-            _reset_config_provider()
-
-    assert mock_write.call_count == 1
-    assert config_path.read_text(encoding="utf-8") == original_text
-
-
 def test_auth_list_does_not_call_mutating_select(monkeypatch, capsys):
     from hermes_cli.auth_commands import auth_list_command
 

tests/hermes_cli/test_auth_nous_provider.py

@@ -76,20 +76,6 @@ class TestResolveVerifyFallback:
         )
         assert result is False
 
-    def test_string_false_in_auth_state_does_not_disable_tls_verify(self):
-        import ssl
-        from hermes_cli.auth import _resolve_verify
-
-        result = _resolve_verify(auth_state={"tls": {"insecure": "false"}})
-        assert result is not False
-        assert result is True or isinstance(result, ssl.SSLContext)
-
-    def test_string_true_in_auth_state_disables_tls_verify(self):
-        from hermes_cli.auth import _resolve_verify
-
-        result = _resolve_verify(auth_state={"tls": {"insecure": "true"}})
-        assert result is False
-
     def test_no_ca_bundle_returns_true(self, monkeypatch):
         from hermes_cli.auth import _resolve_verify
 

tests/hermes_cli/test_commands.py

@@ -405,21 +405,6 @@ class TestGatewayConfigGate:
         joined = "\n".join(lines)
         assert "`/verbose" in joined
 
-    def test_config_gate_quoted_false_stays_disabled_everywhere(self, tmp_path, monkeypatch):
-        """Quoted false must not enable config-gated gateway commands."""
-        config_file = tmp_path / "config.yaml"
-        config_file.write_text('display:\n  tool_progress_command: "false"\n')
-        monkeypatch.setenv("HERMES_HOME", str(tmp_path))
-
-        lines = gateway_help_lines()
-        joined = "\n".join(lines)
-        names = {name for name, _ in telegram_bot_commands()}
-        mapping = slack_subcommand_map()
-
-        assert "`/verbose" not in joined
-        assert "verbose" not in names
-        assert "verbose" not in mapping
-
     def test_config_gate_excluded_from_telegram_when_off(self, tmp_path, monkeypatch):
         config_file = tmp_path / "config.yaml"
         config_file.write_text("display:\n  tool_progress_command: false\n")

tests/hermes_cli/test_goals.py

@@ -1,358 +0,0 @@
-"""Tests for hermes_cli/goals.py — persistent cross-turn goals."""
-
-from __future__ import annotations
-
-import json
-from unittest.mock import patch, MagicMock
-
-import pytest
-
-
-# ──────────────────────────────────────────────────────────────────────
-# Fixtures
-# ──────────────────────────────────────────────────────────────────────
-
-
-@pytest.fixture
-def hermes_home(tmp_path, monkeypatch):
-    """Isolated HERMES_HOME so SessionDB.state_meta writes don't clobber the real one."""
-    from pathlib import Path
-
-    home = tmp_path / ".hermes"
-    home.mkdir()
-    monkeypatch.setattr(Path, "home", lambda: tmp_path)
-    monkeypatch.setenv("HERMES_HOME", str(home))
-
-    # Bust the goal-module's DB cache for each test so it re-resolves HERMES_HOME.
-    from hermes_cli import goals
-
-    goals._DB_CACHE.clear()
-    yield home
-    goals._DB_CACHE.clear()
-
-
-# ──────────────────────────────────────────────────────────────────────
-# _parse_judge_response
-# ──────────────────────────────────────────────────────────────────────
-
-
-class TestParseJudgeResponse:
-    def test_clean_json_done(self):
-        from hermes_cli.goals import _parse_judge_response
-
-        done, reason = _parse_judge_response('{"done": true, "reason": "all good"}')
-        assert done is True
-        assert reason == "all good"
-
-    def test_clean_json_continue(self):
-        from hermes_cli.goals import _parse_judge_response
-
-        done, reason = _parse_judge_response('{"done": false, "reason": "more work needed"}')
-        assert done is False
-        assert reason == "more work needed"
-
-    def test_json_in_markdown_fence(self):
-        from hermes_cli.goals import _parse_judge_response
-
-        raw = '```json\n{"done": true, "reason": "done"}\n```'
-        done, reason = _parse_judge_response(raw)
-        assert done is True
-        assert "done" in reason
-
-    def test_json_embedded_in_prose(self):
-        """Some models prefix reasoning before emitting JSON — we extract it."""
-        from hermes_cli.goals import _parse_judge_response
-
-        raw = 'Looking at this... the agent says X. Verdict: {"done": false, "reason": "partial"}'
-        done, reason = _parse_judge_response(raw)
-        assert done is False
-        assert reason == "partial"
-
-    def test_string_done_values(self):
-        from hermes_cli.goals import _parse_judge_response
-
-        for s in ("true", "yes", "done", "1"):
-            done, _ = _parse_judge_response(f'{{"done": "{s}", "reason": "r"}}')
-            assert done is True
-        for s in ("false", "no", "not yet"):
-            done, _ = _parse_judge_response(f'{{"done": "{s}", "reason": "r"}}')
-            assert done is False
-
-    def test_malformed_json_fails_open(self):
-        """Non-JSON → not done, with error-ish reason (so judge_goal can map to continue)."""
-        from hermes_cli.goals import _parse_judge_response
-
-        done, reason = _parse_judge_response("this is not json at all")
-        assert done is False
-        assert reason  # non-empty
-
-    def test_empty_response(self):
-        from hermes_cli.goals import _parse_judge_response
-
-        done, reason = _parse_judge_response("")
-        assert done is False
-        assert reason
-
-
-# ──────────────────────────────────────────────────────────────────────
-# judge_goal — fail-open semantics
-# ──────────────────────────────────────────────────────────────────────
-
-
-class TestJudgeGoal:
-    def test_empty_goal_skipped(self):
-        from hermes_cli.goals import judge_goal
-
-        verdict, _ = judge_goal("", "some response")
-        assert verdict == "skipped"
-
-    def test_empty_response_continues(self):
-        from hermes_cli.goals import judge_goal
-
-        verdict, _ = judge_goal("ship the thing", "")
-        assert verdict == "continue"
-
-    def test_no_aux_client_continues(self):
-        """Fail-open: if no aux client, we must return continue, not skipped/done."""
-        from hermes_cli import goals
-
-        with patch(
-            "agent.auxiliary_client.get_text_auxiliary_client",
-            return_value=(None, None),
-        ):
-            verdict, _ = goals.judge_goal("my goal", "my response")
-        assert verdict == "continue"
-
-    def test_api_error_continues(self):
-        """Judge exception → fail-open continue (don't wedge progress on judge bugs)."""
-        from hermes_cli import goals
-
-        fake_client = MagicMock()
-        fake_client.chat.completions.create.side_effect = RuntimeError("boom")
-        with patch(
-            "agent.auxiliary_client.get_text_auxiliary_client",
-            return_value=(fake_client, "judge-model"),
-        ):
-            verdict, reason = goals.judge_goal("goal", "response")
-        assert verdict == "continue"
-        assert "judge error" in reason.lower()
-
-    def test_judge_says_done(self):
-        from hermes_cli import goals
-
-        fake_client = MagicMock()
-        fake_client.chat.completions.create.return_value = MagicMock(
-            choices=[
-                MagicMock(
-                    message=MagicMock(content='{"done": true, "reason": "achieved"}')
-                )
-            ]
-        )
-        with patch(
-            "agent.auxiliary_client.get_text_auxiliary_client",
-            return_value=(fake_client, "judge-model"),
-        ):
-            verdict, reason = goals.judge_goal("goal", "agent response")
-        assert verdict == "done"
-        assert reason == "achieved"
-
-    def test_judge_says_continue(self):
-        from hermes_cli import goals
-
-        fake_client = MagicMock()
-        fake_client.chat.completions.create.return_value = MagicMock(
-            choices=[
-                MagicMock(
-                    message=MagicMock(content='{"done": false, "reason": "not yet"}')
-                )
-            ]
-        )
-        with patch(
-            "agent.auxiliary_client.get_text_auxiliary_client",
-            return_value=(fake_client, "judge-model"),
-        ):
-            verdict, reason = goals.judge_goal("goal", "agent response")
-        assert verdict == "continue"
-        assert reason == "not yet"
-
-
-# ──────────────────────────────────────────────────────────────────────
-# GoalManager lifecycle + persistence
-# ──────────────────────────────────────────────────────────────────────
-
-
-class TestGoalManager:
-    def test_no_goal_initial(self, hermes_home):
-        from hermes_cli.goals import GoalManager
-
-        mgr = GoalManager(session_id="test-sid-1")
-        assert mgr.state is None
-        assert not mgr.is_active()
-        assert not mgr.has_goal()
-        assert "No active goal" in mgr.status_line()
-
-    def test_set_then_status(self, hermes_home):
-        from hermes_cli.goals import GoalManager
-
-        mgr = GoalManager(session_id="test-sid-2", default_max_turns=5)
-        state = mgr.set("port the thing")
-        assert state.goal == "port the thing"
-        assert state.status == "active"
-        assert state.max_turns == 5
-        assert state.turns_used == 0
-        assert mgr.is_active()
-        assert "active" in mgr.status_line().lower()
-        assert "port the thing" in mgr.status_line()
-
-    def test_set_rejects_empty(self, hermes_home):
-        from hermes_cli.goals import GoalManager
-
-        mgr = GoalManager(session_id="test-sid-3")
-        with pytest.raises(ValueError):
-            mgr.set("")
-        with pytest.raises(ValueError):
-            mgr.set("   ")
-
-    def test_pause_and_resume(self, hermes_home):
-        from hermes_cli.goals import GoalManager
-
-        mgr = GoalManager(session_id="test-sid-4")
-        mgr.set("goal text")
-        mgr.pause(reason="user-paused")
-        assert mgr.state.status == "paused"
-        assert not mgr.is_active()
-        assert mgr.has_goal()
-
-        mgr.resume()
-        assert mgr.state.status == "active"
-        assert mgr.is_active()
-
-    def test_clear(self, hermes_home):
-        from hermes_cli.goals import GoalManager
-
-        mgr = GoalManager(session_id="test-sid-5")
-        mgr.set("goal")
-        mgr.clear()
-        assert mgr.state is None
-        assert not mgr.is_active()
-
-    def test_persistence_across_managers(self, hermes_home):
-        """Key invariant: a second manager on the same session sees the goal.
-
-        This is what makes /resume work — each session rebinds its
-        GoalManager and picks up the saved state.
-        """
-        from hermes_cli.goals import GoalManager
-
-        mgr1 = GoalManager(session_id="persist-sid")
-        mgr1.set("do the thing")
-
-        mgr2 = GoalManager(session_id="persist-sid")
-        assert mgr2.state is not None
-        assert mgr2.state.goal == "do the thing"
-        assert mgr2.is_active()
-
-    def test_evaluate_after_turn_done(self, hermes_home):
-        """Judge says done → status=done, no continuation."""
-        from hermes_cli import goals
-        from hermes_cli.goals import GoalManager
-
-        mgr = GoalManager(session_id="eval-sid-1")
-        mgr.set("ship it")
-
-        with patch.object(goals, "judge_goal", return_value=("done", "shipped")):
-            decision = mgr.evaluate_after_turn("I shipped the feature.")
-
-        assert decision["verdict"] == "done"
-        assert decision["should_continue"] is False
-        assert decision["continuation_prompt"] is None
-        assert mgr.state.status == "done"
-        assert mgr.state.turns_used == 1
-
-    def test_evaluate_after_turn_continue_under_budget(self, hermes_home):
-        from hermes_cli import goals
-        from hermes_cli.goals import GoalManager
-
-        mgr = GoalManager(session_id="eval-sid-2", default_max_turns=5)
-        mgr.set("a long goal")
-
-        with patch.object(goals, "judge_goal", return_value=("continue", "more work")):
-            decision = mgr.evaluate_after_turn("made some progress")
-
-        assert decision["verdict"] == "continue"
-        assert decision["should_continue"] is True
-        assert decision["continuation_prompt"] is not None
-        assert "a long goal" in decision["continuation_prompt"]
-        assert mgr.state.status == "active"
-        assert mgr.state.turns_used == 1
-
-    def test_evaluate_after_turn_budget_exhausted(self, hermes_home):
-        """When turn budget hits ceiling, auto-pause instead of continuing."""
-        from hermes_cli import goals
-        from hermes_cli.goals import GoalManager
-
-        mgr = GoalManager(session_id="eval-sid-3", default_max_turns=2)
-        mgr.set("hard goal")
-
-        with patch.object(goals, "judge_goal", return_value=("continue", "not yet")):
-            d1 = mgr.evaluate_after_turn("step 1")
-            assert d1["should_continue"] is True
-            assert mgr.state.turns_used == 1
-            assert mgr.state.status == "active"
-
-            d2 = mgr.evaluate_after_turn("step 2")
-            # turns_used is now 2 which equals max_turns → paused
-            assert d2["should_continue"] is False
-            assert mgr.state.status == "paused"
-            assert mgr.state.turns_used == 2
-            assert "budget" in (mgr.state.paused_reason or "").lower()
-
-    def test_evaluate_after_turn_inactive(self, hermes_home):
-        """evaluate_after_turn is a no-op when goal isn't active."""
-        from hermes_cli.goals import GoalManager
-
-        mgr = GoalManager(session_id="eval-sid-4")
-        d = mgr.evaluate_after_turn("anything")
-        assert d["verdict"] == "inactive"
-        assert d["should_continue"] is False
-
-        mgr.set("a goal")
-        mgr.pause()
-        d2 = mgr.evaluate_after_turn("anything")
-        assert d2["verdict"] == "inactive"
-        assert d2["should_continue"] is False
-
-    def test_continuation_prompt_shape(self, hermes_home):
-        """The continuation prompt must include the goal text verbatim —
-        and must be safe to inject as a user-role message (prompt-cache
-        invariants: no system-prompt mutation)."""
-        from hermes_cli.goals import GoalManager
-
-        mgr = GoalManager(session_id="cont-sid")
-        mgr.set("port goal command to hermes")
-        prompt = mgr.next_continuation_prompt()
-        assert prompt is not None
-        assert "port goal command to hermes" in prompt
-        assert prompt.strip()  # non-empty
-
-
-# ──────────────────────────────────────────────────────────────────────
-# Smoke: CommandDef is wired
-# ──────────────────────────────────────────────────────────────────────
-
-
-def test_goal_command_in_registry():
-    from hermes_cli.commands import resolve_command
-
-    cmd = resolve_command("goal")
-    assert cmd is not None
-    assert cmd.name == "goal"
-
-
-def test_goal_command_dispatches_in_cli_registry_helpers():
-    """goal shows up in autocomplete / help categories alongside other Session cmds."""
-    from hermes_cli.commands import COMMANDS, COMMANDS_BY_CATEGORY
-
-    assert "/goal" in COMMANDS
-    session_cmds = COMMANDS_BY_CATEGORY.get("Session", {})
-    assert "/goal" in session_cmds

tests/hermes_cli/test_model_provider_persistence.py

@@ -71,32 +71,6 @@ class TestSaveModelChoiceAlwaysDict:
 
 
 class TestProviderPersistsAfterModelSave:
-    def test_update_config_for_provider_uses_atomic_yaml_write(self, config_home):
-        """Provider switches should delegate config writes to atomic_yaml_write."""
-        from hermes_cli.auth import _update_config_for_provider
-
-        config_path = config_home / "config.yaml"
-        original_text = config_path.read_text(encoding="utf-8")
-
-        def _boom(path, data, **kwargs):
-            assert path == config_path
-            assert data["model"]["provider"] == "nous"
-            assert data["model"]["base_url"] == "https://inference.example.com/v1"
-            assert data["model"]["default"] == "some-old-model"
-            assert kwargs["sort_keys"] is False
-            raise OSError("simulated atomic write failure")
-
-        with patch("hermes_cli.auth.atomic_yaml_write", side_effect=_boom) as mock_write:
-            with pytest.raises(OSError, match="simulated atomic write failure"):
-                _update_config_for_provider(
-                    "nous",
-                    "https://inference.example.com/v1/",
-                    default_model="llama-3.3",
-                )
-
-        assert mock_write.call_count == 1
-        assert config_path.read_text(encoding="utf-8") == original_text
-
     def test_api_key_provider_saved_when_model_was_string(self, config_home, monkeypatch):
         """_model_flow_api_key_provider must persist the provider even when
         config.model started as a plain string."""

tests/hermes_cli/test_plugins.py

@@ -21,7 +21,6 @@ from hermes_cli.plugins import (
     get_plugin_command_handler,
     get_plugin_commands,
     get_pre_tool_call_block_message,
-    resolve_plugin_command_result,
     discover_plugins,
     invoke_hook,
 )
@@ -1062,45 +1061,6 @@ class TestPluginCommands:
         assert mgr._plugin_commands["cmd-b"]["plugin"] == "plugin-b"
 
 
-class TestPluginCommandResultResolution:
-    def test_returns_sync_values_unchanged(self):
-        assert resolve_plugin_command_result("ok") == "ok"
-
-    def test_awaits_async_result_without_running_loop(self):
-        async def _handler():
-            return "async-ok"
-
-        assert resolve_plugin_command_result(_handler()) == "async-ok"
-
-    def test_awaits_async_result_with_running_loop(self, monkeypatch):
-        class _Loop:
-            pass
-
-        async def _handler():
-            return "threaded-ok"
-
-        monkeypatch.setattr("hermes_cli.plugins.asyncio.get_running_loop", lambda: _Loop())
-        assert resolve_plugin_command_result(_handler()) == "threaded-ok"
-
-    def test_running_loop_timeout_does_not_hang_forever(self, monkeypatch):
-        """Threaded path must abort a hung async handler instead of blocking the caller."""
-        import asyncio as _asyncio
-
-        class _Loop:
-            pass
-
-        async def _slow_handler():
-            await _asyncio.sleep(10)
-            return "should-not-reach"
-
-        monkeypatch.setattr("hermes_cli.plugins.asyncio.get_running_loop", lambda: _Loop())
-        monkeypatch.setattr("hermes_cli.plugins._PLUGIN_COMMAND_AWAIT_TIMEOUT_SECS", 0.1)
-
-        import pytest
-        with pytest.raises(TimeoutError):
-            resolve_plugin_command_result(_slow_handler())
-
-
 # ── TestPluginDispatchTool ────────────────────────────────────────────────
 
 

tests/hermes_cli/test_runtime_provider_resolution.py

@@ -897,58 +897,6 @@ def test_named_custom_provider_does_not_shadow_builtin_provider(monkeypatch):
     assert resolved["requested_provider"] == "nous"
 
 
-def test_named_custom_provider_wins_over_builtin_alias(monkeypatch):
-    """A custom_providers entry named after a built-in *alias* (not a canonical
-    provider name) must win over the built-in.  Regression guard for #15743:
-    when users define ``custom_providers: [{name: kimi, ...}]`` and reference
-    ``provider: kimi``, the built-in alias rewriting (``kimi`` → ``kimi-coding``)
-    would otherwise hijack the request and send it to the wrong endpoint.
-    """
-    monkeypatch.setattr(
-        rp,
-        "load_config",
-        lambda: {
-            "custom_providers": [
-                {
-                    "name": "kimi",
-                    "base_url": "https://my-custom-kimi.example.com/v1",
-                    "api_key": "my-kimi-key",
-                }
-            ]
-        },
-    )
-
-    entry = rp._get_named_custom_provider("kimi")
-
-    assert entry is not None
-    assert entry["base_url"] == "https://my-custom-kimi.example.com/v1"
-    assert entry["api_key"] == "my-kimi-key"
-
-
-def test_named_custom_provider_skipped_for_canonical_built_in(monkeypatch):
-    """Companion to the test above: ``nous`` is a canonical provider name
-    (``resolve_provider('nous') == 'nous'``), so a custom entry with that name
-    should NOT be returned — the built-in wins as before.
-    """
-    monkeypatch.setattr(
-        rp,
-        "load_config",
-        lambda: {
-            "custom_providers": [
-                {
-                    "name": "nous",
-                    "base_url": "http://localhost:1234/v1",
-                    "api_key": "shadow-key",
-                }
-            ]
-        },
-    )
-
-    entry = rp._get_named_custom_provider("nous")
-
-    assert entry is None
-
-
 def test_explicit_openrouter_skips_openai_base_url(monkeypatch):
     """When the user explicitly requests openrouter, OPENAI_BASE_URL
     (which may point to a custom endpoint) must not override the

tests/hermes_cli/test_tools_config.py

@@ -120,16 +120,7 @@ def test_get_platform_tools_preserves_explicit_empty_selection():
 
     enabled = _get_platform_tools(config, "cli")
 
-    # An explicit empty list disables every CONFIGURABLE toolset (web,
-    # terminal, memory, …). Non-configurable platform toolsets that ride
-    # along on the platform's default composite (e.g. `kanban`, whose tools
-    # live in _HERMES_CORE_TOOLS but aren't user-toggleable) are still
-    # auto-recovered by _get_platform_tools so saving via `hermes tools`
-    # doesn't silently drop them. The contract this test guards is the
-    # configurable side: nothing the user could have checked in the TUI
-    # checklist should reappear here.
-    configurable = {ts_key for ts_key, _, _ in CONFIGURABLE_TOOLSETS}
-    assert enabled.isdisjoint(configurable)
+    assert enabled == set()
 
 
 def test_apply_toolset_change_from_default_does_not_enable_default_off_toolsets():

tests/hermes_cli/test_update_gateway_restart.py

@@ -392,81 +392,6 @@ class TestCmdUpdateLaunchdRestart:
         captured = capsys.readouterr().out
         assert "Restart manually: hermes gateway run" in captured
 
-    @patch("shutil.which", return_value=None)
-    @patch("subprocess.run")
-    def test_update_restarts_profile_manual_gateways(
-        self, mock_run, _mock_which, mock_args, capsys, tmp_path, monkeypatch,
-    ):
-        """Profile-mapped manual gateways are relaunched automatically after update."""
-        monkeypatch.setattr(gateway_cli, "is_macos", lambda: True)
-        monkeypatch.setattr(
-            gateway_cli,
-            "get_launchd_plist_path",
-            lambda: tmp_path / "ai.hermes.gateway.plist",
-        )
-
-        mock_run.side_effect = _make_run_side_effect(
-            commit_count="3",
-            launchctl_loaded=False,
-        )
-        process = gateway_cli.ProfileGatewayProcess(
-            profile="coder",
-            path=tmp_path / ".hermes" / "profiles" / "coder",
-            pid=12345,
-        )
-
-        with patch.object(gateway_cli, "find_gateway_pids", return_value=[12345]), \
-             patch.object(gateway_cli, "find_profile_gateway_processes", return_value=[process]), \
-             patch.object(gateway_cli, "launch_detached_profile_gateway_restart", return_value=True) as restart, \
-             patch.object(gateway_cli, "_graceful_restart_via_sigusr1", return_value=True) as graceful, \
-             patch("os.kill") as kill:
-            cmd_update(mock_args)
-
-        captured = capsys.readouterr().out
-        restart.assert_called_once_with("coder", 12345)
-        graceful.assert_called_once()
-        # Graceful drain succeeded — no SIGTERM fallback needed.
-        kill.assert_not_called()
-        assert "Restarting manual gateway profile(s): coder" in captured
-        assert "Restart manually: hermes gateway run" not in captured
-
-    @patch("shutil.which", return_value=None)
-    @patch("subprocess.run")
-    def test_update_profile_manual_gateway_falls_back_to_sigterm(
-        self, mock_run, _mock_which, mock_args, capsys, tmp_path, monkeypatch,
-    ):
-        """When graceful SIGUSR1 drain fails, manual profile restart falls back to SIGTERM."""
-        monkeypatch.setattr(gateway_cli, "is_macos", lambda: True)
-        monkeypatch.setattr(
-            gateway_cli,
-            "get_launchd_plist_path",
-            lambda: tmp_path / "ai.hermes.gateway.plist",
-        )
-
-        mock_run.side_effect = _make_run_side_effect(
-            commit_count="3",
-            launchctl_loaded=False,
-        )
-        process = gateway_cli.ProfileGatewayProcess(
-            profile="coder",
-            path=tmp_path / ".hermes" / "profiles" / "coder",
-            pid=12345,
-        )
-
-        with patch.object(gateway_cli, "find_gateway_pids", return_value=[12345]), \
-             patch.object(gateway_cli, "find_profile_gateway_processes", return_value=[process]), \
-             patch.object(gateway_cli, "launch_detached_profile_gateway_restart", return_value=True) as restart, \
-             patch.object(gateway_cli, "_graceful_restart_via_sigusr1", return_value=False) as graceful, \
-             patch("os.kill") as kill:
-            cmd_update(mock_args)
-
-        captured = capsys.readouterr().out
-        restart.assert_called_once_with("coder", 12345)
-        graceful.assert_called_once()
-        # Graceful drain returned False → SIGTERM fallback.
-        kill.assert_called_once()
-        assert "Restarting manual gateway profile(s): coder" in captured
-
     @patch("shutil.which", return_value=None)
     @patch("subprocess.run")
     def test_update_with_systemd_still_restarts_via_systemd(