fix(gateway): keep event loop responsive (NAN-33)

Move blocking filesystem, persistence, subprocess, media, and DNS work off the gateway event loop while preserving existing contracts. Add bounded cancellation and responsiveness regression coverage.

nanobot/agent/autocompact.py

@@ -2,7 +2,9 @@
 
 from __future__ import annotations
 
-from collections.abc import Collection
+import asyncio
+import inspect
+from collections.abc import Awaitable, Collection
 from datetime import datetime
 from typing import TYPE_CHECKING, Any, Callable, Coroutine
 
@@ -47,9 +49,27 @@ class AutoCompact:
         return idle_seconds >= self._ttl * 60
 
     def _has_unarchived_messages(self, key: str) -> bool:
-        session = self.sessions.get_or_create(key)
+        return self._session_has_unarchived_messages(self.sessions.get_or_create(key))
+
+    @staticmethod
+    def _session_has_unarchived_messages(session: Session) -> bool:
         return session.last_consolidated < len(session.messages)
 
+    def _has_native_async_session_method(self, name: str) -> bool:
+        """Check the manager's real class, not mock-generated instance attributes."""
+        method = inspect.getattr_static(type(self.sessions), name, None)
+        return inspect.iscoroutinefunction(method)
+
+    async def _list_sessions_nonblocking(self) -> list[dict[str, Any]]:
+        if self._has_native_async_session_method("list_sessions_async"):
+            return await self.sessions.list_sessions_async()
+        return await asyncio.to_thread(self.sessions.list_sessions)
+
+    async def _get_or_create_nonblocking(self, key: str) -> Session:
+        if self._has_native_async_session_method("get_or_create_async"):
+            return await self.sessions.get_or_create_async(key)
+        return await asyncio.to_thread(self.sessions.get_or_create, key)
+
     @classmethod
     def _is_internal_session(cls, key: str) -> bool:
         return key.startswith(cls._INTERNAL_SESSION_PREFIXES)
@@ -79,6 +99,31 @@ class AutoCompact:
                 self._archiving.add(key)
                 schedule_background(self._archive(key, runtime=runtime))
 
+    async def check_expired_async(
+        self,
+        schedule_background: Callable[[Coroutine[Any, Any, None]], None],
+        resolve_runtime: Callable[[Session], Awaitable[LLMRuntime]],
+        active_session_keys: Collection[str] = (),
+    ) -> None:
+        """Schedule idle archival without blocking the event loop."""
+        now = datetime.now()
+        active_keys = set(active_session_keys)
+        for info in await self._list_sessions_nonblocking():
+            key = info.get("key", "")
+            if not key or self._is_internal_session(key) or key in self._archiving:
+                continue
+            if key in active_keys or not self._is_expired(info.get("updated_at"), now):
+                continue
+            session = await self._get_or_create_nonblocking(key)
+            if not self._session_has_unarchived_messages(session):
+                continue
+            try:
+                runtime = await resolve_runtime(session)
+            except (KeyError, ValueError):
+                continue
+            self._archiving.add(key)
+            schedule_background(self._archive_async(key, runtime=runtime))
+
     async def _archive(self, key: str, *, runtime: LLMRuntime) -> None:
         if self._is_internal_session(key):
             self._archiving.discard(key)
@@ -90,17 +135,37 @@ class AutoCompact:
                 max_suffix=self._RECENT_SUFFIX_MESSAGES,
             )
             if summary and summary != "(nothing)":
-                session = self.sessions.get_or_create(key)
+                self._record_stored_summary(key, self.sessions.get_or_create(key))
+        except Exception:
+            logger.exception("Auto-compact: failed for {}", key)
+        finally:
+            self._archiving.discard(key)
+
+    async def _archive_async(self, key: str, *, runtime: LLMRuntime) -> None:
+        if self._is_internal_session(key):
+            self._archiving.discard(key)
+            return
+        try:
+            summary = await self.consolidator.compact_idle_session(
+                key,
+                runtime=runtime,
+                max_suffix=self._RECENT_SUFFIX_MESSAGES,
+            )
+            if summary and summary != "(nothing)":
+                session = await self._get_or_create_nonblocking(key)
+                self._record_stored_summary(key, session)
+        except Exception:
+            logger.exception("Auto-compact: failed for {}", key)
+        finally:
+            self._archiving.discard(key)
+
+    def _record_stored_summary(self, key: str, session: Session) -> None:
         stored = session_summary_from_metadata(
             session.metadata,
             fallback_last_active=session.updated_at,
         )
         if stored is not None:
             self._summaries[key] = stored
-        except Exception:
-            logger.exception("Auto-compact: failed for {}", key)
-        finally:
-            self._archiving.discard(key)
 
     def prepare_session(self, session: Session, key: str) -> tuple[Session, SessionSummary | None]:
         if self._is_internal_session(key):
@@ -110,6 +175,28 @@ class AutoCompact:
         if key in self._archiving or self._is_expired(session.updated_at):
             logger.info("Auto-compact: reloading session {} (archiving={})", key, key in self._archiving)
             session = self.sessions.get_or_create(key)
+        return self._prepared_summary(session, key)
+
+    async def prepare_session_async(
+        self,
+        session: Session,
+        key: str,
+    ) -> tuple[Session, SessionSummary | None]:
+        """Prepare a session without blocking on a reload."""
+        if self._is_internal_session(key):
+            self._archiving.discard(key)
+            self._summaries.pop(key, None)
+            return session, None
+        if key in self._archiving or self._is_expired(session.updated_at):
+            logger.info("Auto-compact: reloading session {} (archiving={})", key, key in self._archiving)
+            session = await self._get_or_create_nonblocking(key)
+        return self._prepared_summary(session, key)
+
+    def _prepared_summary(
+        self,
+        session: Session,
+        key: str,
+    ) -> tuple[Session, SessionSummary | None]:
         # Hot path: summary from in-memory dict (process hasn't restarted).
         entry = self._summaries.pop(key, None)
         if entry:

nanobot/agent/automation_turns.py

@@ -13,6 +13,19 @@ class AutomationTurnError(RuntimeError):
     """Raised when an automation turn reaches the agent and finishes with an error."""
 
 
+class AutomationTurnAcceptedCancellation(asyncio.CancelledError):
+    """Cancellation raised after an automation turn was accepted for processing.
+
+    Callers must not replay the turn: the accepted agent work now has independent
+    ownership and may continue after the submitting task is cancelled.
+    """
+
+
+def _consume_future_exception(future: asyncio.Future[object]) -> None:
+    if not future.cancelled():
+        future.exception()
+
+
 async def publish_next_deferred_turn(
     *,
     deferred_queues: dict[str, list[InboundMessage]],
@@ -70,19 +83,36 @@ class AutomationTurnCoordinator:
         future: asyncio.Future[OutboundMessage | None] = loop.create_future()
         self._waiters[turn_id] = future
         self._pending_messages_by_turn_id[turn_id] = msg
+        accepted = False
         try:
             if self._is_running():
                 await self._publish_inbound(msg)
+                accepted = True
             else:
-                await self._dispatch(msg)
+                # Direct dispatch is given independent task ownership for the
+                # same reason as publishing to the inbound queue: once admitted,
+                # cancelling this submitter must not cancel and then replay the
+                # already-running agent turn.
+                dispatch_future: asyncio.Future[object] = asyncio.ensure_future(
+                    self._dispatch(msg)
+                )
+                dispatch_future.add_done_callback(_consume_future_exception)
+                accepted = True
+                await asyncio.shield(dispatch_future)
             try:
                 return await future
-            except asyncio.CancelledError:
+            except asyncio.CancelledError as exc:
+                if accepted:
+                    raise AutomationTurnAcceptedCancellation(*exc.args) from None
                 raise
             except AutomationTurnError:
                 raise
             except Exception as exc:
                 raise AutomationTurnError(str(exc) or exc.__class__.__name__) from exc
+        except asyncio.CancelledError as exc:
+            if accepted and not isinstance(exc, AutomationTurnAcceptedCancellation):
+                raise AutomationTurnAcceptedCancellation(*exc.args) from None
+            raise
         finally:
             self._waiters.pop(turn_id, None)
             self._pending_messages_by_turn_id.pop(turn_id, None)

nanobot/agent/loop.py

@@ -67,6 +67,7 @@ from nanobot.security.workspace_access import (
     reset_workspace_scope,
 )
 from nanobot.session import turn_continuation
+from nanobot.session.async_compat import call_session_manager
 from nanobot.session.automation_turns import automation_history_overrides
 from nanobot.session.goal_state import (
     goal_state_runtime_lines,
@@ -540,6 +541,33 @@ class AgentLoop:
             **extra,
         )
 
+    async def _get_or_create_session(self, key: str) -> Session:
+        """Use native async session loading, with a compatibility fallback."""
+        return await call_session_manager(
+            self.sessions,
+            "get_or_create_async",
+            self.sessions.get_or_create,
+            key,
+        )
+
+    async def _save_session(self, session: Session) -> None:
+        """Use native async session saving, with a compatibility fallback."""
+        await call_session_manager(
+            self.sessions,
+            "save_async",
+            self.sessions.save,
+            session,
+        )
+
+    async def _save_runtime_checkpoint(self, session: Session) -> None:
+        """Use native async checkpoint saving, with a compatibility fallback."""
+        await call_session_manager(
+            self.sessions,
+            "save_runtime_checkpoint_async",
+            self.sessions.save_runtime_checkpoint,
+            session,
+        )
+
     def _sync_subagent_runtime_limits(self) -> None:
         """Keep subagent runtime limits aligned with mutable loop settings."""
         self.subagents.max_iterations = self.max_iterations
@@ -579,6 +607,30 @@ class AgentLoop:
             self.sessions.save(session)
             return self.llm_runtime()
 
+    async def runtime_for_session_async(
+        self,
+        session: Session,
+        *,
+        recover_removed: bool = True,
+    ) -> LLMRuntime:
+        """Resolve a session runtime without blocking on recovery persistence."""
+        name = model_preset_from_metadata(session.metadata)
+        if name is None:
+            return self.llm_runtime()
+        try:
+            return self.runtime_resolver.resolve_preset(name)
+        except KeyError:
+            if not recover_removed or name in self.runtime_resolver.model_presets:
+                raise
+            logger.warning(
+                "Session '{}' references removed model preset '{}'; falling back to default",
+                session.key,
+                name,
+            )
+            session.metadata.pop(SESSION_MODEL_PRESET_METADATA_KEY, None)
+            await self._save_session(session)
+            return self.llm_runtime()
+
     def set_session_model_preset(
         self,
         session_key: str,
@@ -591,6 +643,18 @@ class AgentLoop:
         self.sessions.save(session)
         return runtime
 
+    async def set_session_model_preset_async(
+        self,
+        session_key: str,
+        name: str,
+    ) -> LLMRuntime:
+        """Validate and persist one session's preset selection without blocking."""
+        runtime = self.runtime_resolver.resolve_preset(name)
+        session = await self._get_or_create_session(session_key)
+        session.metadata[SESSION_MODEL_PRESET_METADATA_KEY] = runtime.model_preset
+        await self._save_session(session)
+        return runtime
+
     def _publish_runtime_selection(
         self,
         runtime: LLMRuntime,
@@ -703,17 +767,14 @@ class AgentLoop:
             session_key=session_key,
         )
 
-    def _persist_user_message_early(
+    def _stage_user_message_early(
         self,
         msg: InboundMessage,
         session: Session,
         runtime_context_blocks: list[RuntimeContextBlock] | None = None,
         **kwargs: Any,
     ) -> bool:
-        """Persist the triggering user message before the turn starts.
+        """Add the triggering user message and recovery markers in memory."""
-
-        Returns True if the message was persisted.
-        """
         if not turn_continuation.should_persist_user_message(msg.metadata):
             return False
         media_paths = [
@@ -742,10 +803,45 @@ class AgentLoop:
             followup_id = msg.metadata.get(PENDING_FOLLOWUP_ID_KEY)
             if isinstance(followup_id, str) and followup_id:
                 acknowledge_pending_followups(session, [followup_id])
-            self.sessions.save(session)
             return True
         return False
 
+    def _persist_user_message_early(
+        self,
+        msg: InboundMessage,
+        session: Session,
+        runtime_context_blocks: list[RuntimeContextBlock] | None = None,
+        **kwargs: Any,
+    ) -> bool:
+        """Synchronously persist the user message for compatibility callers."""
+        persisted = self._stage_user_message_early(
+            msg,
+            session,
+            runtime_context_blocks,
+            **kwargs,
+        )
+        if persisted:
+            self.sessions.save(session)
+        return persisted
+
+    async def _persist_user_message_early_async(
+        self,
+        msg: InboundMessage,
+        session: Session,
+        runtime_context_blocks: list[RuntimeContextBlock] | None = None,
+        **kwargs: Any,
+    ) -> bool:
+        """Persist the user message without blocking the event loop."""
+        persisted = self._stage_user_message_early(
+            msg,
+            session,
+            runtime_context_blocks,
+            **kwargs,
+        )
+        if persisted:
+            await self._save_session(session)
+        return persisted
+
     def _build_initial_messages(self, ctx: TurnContext) -> list[dict[str, Any]]:
         """Build the initial message list for the LLM turn."""
         assert ctx.session is not None
@@ -843,7 +939,7 @@ class AgentLoop:
         if tool is None:
             content = "Shell execution is disabled in this nanobot configuration."
         else:
-            session = ctx.session or self.sessions.get_or_create(ctx.key)
+            session = ctx.session or await AgentLoop._get_or_create_session(self, ctx.key)
             scope = self.workspace_scopes.for_turn(
                 channel=ctx.msg.channel,
                 message_metadata=metadata,
@@ -1001,7 +1097,7 @@ class AgentLoop:
                 public_payload[self._PROVIDER_STATE_CHECKPOINT_VERSION_KEY] = (
                     self._PROVIDER_STATE_CHECKPOINT_VERSION
                 )
-            self._set_runtime_checkpoint(session, public_payload)
+            await self._set_runtime_checkpoint_async(session, public_payload)
 
         async def _drain_pending(*, limit: int = _MAX_INJECTIONS_PER_TURN) -> list[dict[str, Any]]:
             """Drain follow-up messages from the pending queue.
@@ -1239,18 +1335,33 @@ class AgentLoop:
             logger.error("LLM returned error: {}", (result.final_content or "")[:200])
         return result.final_content, result.tools_used, result.messages, result.stop_reason, result.had_injections
 
-    def _check_expired_sessions_if_due(self) -> None:
+    def _idle_compact_scan_due(self) -> bool:
-        """Scan idle sessions no more often than the configured interval."""
         now = time.monotonic()
         if now < self._next_idle_compact_check_at:
-            return
+            return False
         self._next_idle_compact_check_at = now + self._idle_compact_check_interval_s
+        return True
+
+    def _check_expired_sessions_if_due(self) -> None:
+        """Synchronously scan idle sessions for compatibility with direct callers."""
+        if not self._idle_compact_scan_due():
+            return
         self.auto_compact.check_expired(
             self.schedule_background,
             self.runtime_for_session,
             active_session_keys=self._pending_queues.keys(),
         )
 
+    async def _check_expired_sessions_if_due_async(self) -> None:
+        """Scan idle sessions without blocking the event loop."""
+        if not self._idle_compact_scan_due():
+            return
+        await self.auto_compact.check_expired_async(
+            self.schedule_background,
+            self.runtime_for_session_async,
+            active_session_keys=self._pending_queues.keys(),
+        )
+
     async def run(self) -> None:
         """Run the agent loop, dispatching messages as tasks to stay responsive to /stop."""
         self._running = True
@@ -1261,7 +1372,7 @@ class AgentLoop:
                 try:
                     msg = await asyncio.wait_for(self.bus.consume_inbound(), timeout=1.0)
                 except asyncio.TimeoutError:
-                    self._check_expired_sessions_if_due()
+                    await self._check_expired_sessions_if_due_async()
                     continue
                 except asyncio.CancelledError:
                     # Preserve real task cancellation so shutdown can complete cleanly.
@@ -1339,7 +1450,7 @@ class AgentLoop:
                         )
                         continue
                     pending_msg = routed_msg
-                    session = self.sessions.get_or_create(effective_key)
+                    session = await self._get_or_create_session(effective_key)
                     followup_id = record_pending_followup(session, pending_msg)
                     if followup_id is not None:
                         pending_msg = dataclasses.replace(
@@ -1349,7 +1460,7 @@ class AgentLoop:
                                 PENDING_FOLLOWUP_ID_KEY: followup_id,
                             },
                         )
-                        self.sessions.save(session)
+                        await self._save_session(session)
                     try:
                         self._pending_queues[effective_key].put_nowait(pending_msg)
                     except asyncio.QueueFull:
@@ -1460,10 +1571,10 @@ class AgentLoop:
                         raise
                     try:
                         key = self._effective_session_key(msg)
-                        session = self.sessions.get_or_create(key)
+                        session = await self._get_or_create_session(key)
                         if self._restore_runtime_checkpoint(session):
                             self._clear_pending_user_turn(session)
-                            self.sessions.save(session)
+                            await self._save_session(session)
                             logger.info(
                                 "Restored partial context for cancelled session {}",
                                 key,
@@ -1788,7 +1899,7 @@ class AgentLoop:
                 if ctx.session is None:
                     raise RuntimeError("required session is not active")
             else:
-                ctx.session = self.sessions.get_or_create(ctx.session_key)
+                ctx.session = await self._get_or_create_session(ctx.session_key)
         session = ctx.session
         ctx.ephemeral = ctx.ephemeral or not session.policy.persist
         tools = ctx.tools or self.tools
@@ -1819,16 +1930,16 @@ class AgentLoop:
             self.workspace_scopes.persist_message_scope(session, msg)
 
         if self._restore_runtime_checkpoint(session):
-            self.sessions.save(session)
+            await self._save_session(session)
         if (
             RECOVERY_INBOUND_METADATA_KEY not in msg.metadata
             and restore_pending_interruption(session)
         ):
-            self.sessions.save(session)
+            await self._save_session(session)
 
     async def _compact_session(self, ctx: TurnContext) -> None:
         session = ctx.require_session()
-        ctx.session, pending = self.auto_compact.prepare_session(
+        ctx.session, pending = await self.auto_compact.prepare_session_async(
             session,
             ctx.session_key,
         )
@@ -1865,14 +1976,14 @@ class AgentLoop:
             # them out of LLM context.  /new is excluded because it
             # intentionally clears the session.
             if cmd_ctx.raw.lower() != "/new":
-                ctx.input_persisted_early = self._persist_user_message_early(
+                ctx.input_persisted_early = await self._persist_user_message_early_async(
                     ctx.msg, session, _command=True
                 )
                 session.add_message(
                     "assistant", result.content, _command=True
                 )
                 self._clear_pending_user_turn(session)
-                self.sessions.save(session)
+                await self._save_session(session)
                 if not ctx.ephemeral:
                     await self.runtime_event_publisher.session_turn_persisted(
                         ctx.msg,
@@ -1887,7 +1998,7 @@ class AgentLoop:
         session = ctx.require_session()
         runtime = ctx.runtime
         if runtime is None:
-            runtime = self.runtime_for_session(session)
+            runtime = await self.runtime_for_session_async(session)
             ctx.runtime = runtime
         if ctx.session_key.startswith("dream:"):
             logger.info(
@@ -1931,7 +2042,7 @@ class AgentLoop:
                 # provider compatibility or prompt assembly work. A compatible
                 # staged state replaces this in a second atomic save below.
                 session.provider_state = None
-                self.sessions.save(session)
+                await self._save_session(session)
             ctx.input_persisted_early = True
         await ctx.delivery.runtime_admitted(runtime)
 
@@ -1985,7 +2096,7 @@ class AgentLoop:
         elif stored_state is not None:
             session.provider_state = None
         if ctx.kind is TurnKind.USER:
-            ctx.input_persisted_early = self._persist_user_message_early(
+            ctx.input_persisted_early = await self._persist_user_message_early_async(
                 ctx.msg,
                 session,
                 runtime_context_blocks=ctx.runtime_context_blocks,
@@ -1995,7 +2106,7 @@ class AgentLoop:
         elif subagent_followup_persisted and staged_provider_state:
             # Upgrade the replay-safe baseline to the resumable state before
             # prompt assembly and the first model checkpoint.
-            self.sessions.save(session)
+            await self._save_session(session)
         ctx.initial_messages = self._build_initial_messages(ctx)
 
         if ctx.on_progress is None:
@@ -2071,6 +2182,9 @@ class AgentLoop:
             turn_latency_ms=ctx.turn_latency_ms,
         )
         ctx.delivery.record_latency(ctx.turn_latency_ms)
+        self._clear_pending_user_turn(session)
+        self._clear_runtime_checkpoint(session)
+        await self._save_session(session)
         if not ctx.ephemeral:
             self.schedule_background(
                 self.consolidator.maybe_consolidate_by_tokens(
@@ -2078,10 +2192,6 @@ class AgentLoop:
                     runtime=runtime,
                 )
             )
-        self._clear_pending_user_turn(session)
-        self._clear_runtime_checkpoint(session)
-        self.sessions.save(session)
-        if not ctx.ephemeral:
             await self.runtime_event_publisher.session_turn_persisted(
                 ctx.msg,
                 ctx.session_key,
@@ -2294,11 +2404,24 @@ class AgentLoop:
         )
         return True
 
-    def _set_runtime_checkpoint(self, session: Session, payload: dict[str, Any]) -> None:
+    def _set_runtime_checkpoint(
-        """Persist the latest in-flight turn state into session metadata."""
+        self,
+        session: Session,
+        payload: dict[str, Any],
+    ) -> None:
+        """Synchronously persist a checkpoint for compatibility callers."""
         session.metadata[self._RUNTIME_CHECKPOINT_KEY] = payload
         self.sessions.save_runtime_checkpoint(session)
 
+    async def _set_runtime_checkpoint_async(
+        self,
+        session: Session,
+        payload: dict[str, Any],
+    ) -> None:
+        """Persist the latest in-flight turn state without blocking the event loop."""
+        session.metadata[self._RUNTIME_CHECKPOINT_KEY] = payload
+        await self._save_runtime_checkpoint(session)
+
     def _mark_pending_user_turn(self, session: Session) -> None:
         session.metadata[self._PENDING_USER_TURN_KEY] = True
 

nanobot/agent/memory.py

@@ -22,6 +22,7 @@ from loguru import logger
 
 from nanobot.llm_usage.context import llm_usage_source
 from nanobot.runtime_context import public_history_messages
+from nanobot.session.async_compat import call_session_manager
 from nanobot.session.manager import (
     MIN_COMPACTED_REPLAY_MESSAGES,
     Session,
@@ -824,6 +825,22 @@ class Consolidator:
             weakref.WeakValueDictionary()
         )
 
+    async def _get_or_create_session(self, key: str) -> Session:
+        return await call_session_manager(
+            self.sessions,
+            "get_or_create_async",
+            self.sessions.get_or_create,
+            key,
+        )
+
+    async def _save_session(self, session: Session) -> None:
+        await call_session_manager(
+            self.sessions,
+            "save_async",
+            self.sessions.save,
+            session,
+        )
+
     def get_lock(self, session_key: str) -> asyncio.Lock:
         """Return the shared consolidation lock for one session."""
         return self._locks.setdefault(session_key, asyncio.Lock())
@@ -859,13 +876,13 @@ class Consolidator:
             return []
         return session.get_history()
 
-    def _persist_last_summary(self, session: Session, summary: str | None) -> None:
+    async def _persist_last_summary(self, session: Session, summary: str | None) -> None:
         if summary and summary != "(nothing)":
             session.metadata["_last_summary"] = {
                 "text": summary,
                 "last_active": session.updated_at.isoformat(),
             }
-            self.sessions.save(session)
+            await self._save_session(session)
 
     def estimate_session_prompt_tokens(
         self,
@@ -1057,7 +1074,7 @@ class Consolidator:
         lock = self.get_lock(session.key)
         async with lock:
             # Refresh session reference: AutoCompact may have replaced it.
-            fresh = self.sessions.get_or_create(session.key)
+            fresh = await self._get_or_create_session(session.key)
             if fresh is not session:
                 session = fresh
             if not session.messages:
@@ -1071,7 +1088,7 @@ class Consolidator:
                 runtime=runtime,
             )
             if estimated <= 0:
-                self._persist_last_summary(session, last_summary)
+                await self._persist_last_summary(session, last_summary)
                 return
             if estimated < budget:
                 unconsolidated_count = len(session.messages) - session.last_consolidated
@@ -1083,7 +1100,7 @@ class Consolidator:
                     source,
                     unconsolidated_count,
                 )
-                self._persist_last_summary(session, last_summary)
+                await self._persist_last_summary(session, last_summary)
                 return
 
             for round_num in range(self._MAX_CONSOLIDATION_ROUNDS):
@@ -1127,7 +1144,7 @@ class Consolidator:
                     last_summary = summary
                 session.last_consolidated = end_idx
                 session.provider_state = None
-                self.sessions.save(session)
+                await self._save_session(session)
                 if not summary:
                     # LLM is degraded — stop hammering it this call;
                     # the next invocation can retry a fresh chunk.
@@ -1143,7 +1160,7 @@ class Consolidator:
             # Persist the last summary to session metadata so it can be injected
             # into the runtime context on the next prepare_session() call, aligning
             # the summary injection strategy with AutoCompact._archive().
-            self._persist_last_summary(session, last_summary)
+            await self._persist_last_summary(session, last_summary)
 
     async def compact_idle_session(
         self,
@@ -1168,7 +1185,7 @@ class Consolidator:
         lock = self.get_lock(session_key)
         async with lock:
             self.sessions.invalidate(session_key)
-            session = self.sessions.get_or_create(session_key)
+            session = await self._get_or_create_session(session_key)
 
             archive_start = session.last_consolidated
             messages_to_archive = list(session.messages[archive_start:])
@@ -1193,7 +1210,7 @@ class Consolidator:
             # through the captured batch so new messages remain eligible next time.
             session.last_consolidated = archive_end
             session.provider_state = None
-            self.sessions.save(session)
+            await self._save_session(session)
 
             visible = session.get_history(
                 max_messages=MIN_COMPACTED_REPLAY_MESSAGES,

nanobot/agent/runner.py

@@ -6,7 +6,7 @@ import asyncio
 import inspect
 import os
 import time
-from collections.abc import Awaitable, Callable, Iterable
+from collections.abc import Awaitable, Callable, Iterable, Sized
 from copy import deepcopy
 from dataclasses import dataclass, field
 from pathlib import Path
@@ -83,6 +83,22 @@ _MAX_EMPTY_RETRIES = 2
 _MAX_LENGTH_RECOVERIES = 3
 _MAX_INJECTIONS_PER_TURN = 3
 _MAX_INJECTION_CYCLES = 5
+_SLOW_TOOL_LOG_MS = 1_000
+
+
+def _tool_input_scale(params: object) -> tuple[int, int]:
+    """Return bounded structural counts without logging argument content."""
+    if not isinstance(params, dict):
+        return 0, len(params) if isinstance(params, str | bytes) else 0
+    params_dict = cast(dict[object, object], params)
+    items = len(params_dict)
+    chars = 0
+    for value in params_dict.values():
+        if isinstance(value, str | bytes):
+            chars += len(value)
+        elif isinstance(value, list | tuple | set | dict):
+            items += len(cast(Sized, value))
+    return items, chars
 
 
 def _restore_outer_whitespace(content: str, original: str | None) -> str:
@@ -1521,6 +1537,7 @@ class AgentRunner:
                 RuntimeError(prep_error) if spec.fail_on_tool_error else None
             )
         await hook.before_execute_tool(context, tool_call, tool, params)
+        tool_started_at = time.perf_counter()
         try:
             if tool is not None:
                 result = await tool.execute(**params)
@@ -1549,6 +1566,17 @@ class AgentRunner:
             if spec.fail_on_tool_error:
                 return payload, event, exc
             return payload, event, None
+        finally:
+            duration_ms = int((time.perf_counter() - tool_started_at) * 1000)
+            if duration_ms >= _SLOW_TOOL_LOG_MS:
+                input_items, input_chars = _tool_input_scale(params)
+                logger.warning(
+                    "slow tool operation={} input_items={} input_chars={} duration_ms={}",
+                    tool_call.name,
+                    input_items,
+                    input_chars,
+                    duration_ms,
+                )
 
         if is_tool_error_result(result):
             await hook.on_execute_tool_error(context, tool_call, tool, params, result)

nanobot/agent/tools/cli_apps.py

@@ -4,7 +4,10 @@
 
 from __future__ import annotations
 
+import asyncio
+import inspect
 from pathlib import Path
+from typing import TypedDict
 
 from pydantic import Field
 
@@ -24,6 +27,14 @@ from nanobot.runtime_context import RuntimeContextBlock, wrap_runtime_context_li
 from nanobot.security.workspace_access import current_tool_workspace
 
 
+class _CliAppRunKwargs(TypedDict):
+    args: list[str]
+    json_output: bool
+    working_dir: str | None
+    timeout: int | None
+    restrict_to_workspace: bool
+
+
 class CliAppsToolConfig(Base):
     """CLI Apps tool configuration."""
 
@@ -147,14 +158,17 @@ class CliAppsTool(Tool):
         )
         workspace = access.project_path or self.workspace
         manager = CliAppManager(workspace=workspace, runtime=self.runtime)
+        run_kwargs: _CliAppRunKwargs = {
+            "args": args or [],
+            "json_output": bool(json),
+            "working_dir": working_dir,
+            "timeout": timeout,
+            "restrict_to_workspace": access.restrict_to_workspace,
+        }
         try:
-            return manager.run(
+            run_async = inspect.getattr_static(type(manager), "run_async", None)
-                name,
+            if inspect.iscoroutinefunction(run_async):
-                args=args or [],
+                return await manager.run_async(name, **run_kwargs)
-                json_output=bool(json),
+            return await asyncio.to_thread(manager.run, name, **run_kwargs)
-                working_dir=working_dir,
-                timeout=timeout,
-                restrict_to_workspace=access.restrict_to_workspace,
-            )
         except CliAppError as exc:
             return ToolResult.error(f"Error: {exc.message}")

nanobot/agent/tools/cron.py

@@ -143,14 +143,41 @@ class CronTool(Tool):
         tz: str | None = None,
         at: str | None = None,
         job_id: str | None = None,
+    ) -> str:
+        if action == "add" and self._in_cron_context.get():
+            return ToolResult.error(
+                "Error: cannot schedule new jobs from within a cron job execution"
+            )
+        return await self._cron.run_sync(
+            self._execute_sync,
+            action,
+            name,
+            message,
+            every_seconds,
+            cron_expr,
+            tz,
+            at,
+            job_id,
+        )
+
+    def _execute_sync(
+        self,
+        action: str,
+        name: str | None,
+        message: str,
+        every_seconds: int | None,
+        cron_expr: str | None,
+        tz: str | None,
+        at: str | None,
+        job_id: str | None,
     ) -> str:
         if action == "add":
             if self._in_cron_context.get():
                 return ToolResult.error("Error: cannot schedule new jobs from within a cron job execution")
             return self._add_job(name, message, every_seconds, cron_expr, tz, at)
-        elif action == "list":
+        if action == "list":
             return self._list_jobs()
-        elif action == "remove":
+        if action == "remove":
             return self._remove_job(job_id)
         return f"Unknown action: {action}"
 

nanobot/agent/tools/filesystem.py

@@ -2,9 +2,11 @@
 
 # pyright: reportPrivateUsage=false, reportUnusedFunction=false
 
+import asyncio
 import difflib
 import mimetypes
 import os
+import threading
 from dataclasses import dataclass
 from pathlib import Path
 from typing import Any
@@ -21,6 +23,7 @@ from nanobot.agent.tools.schema import (
 )
 from nanobot.config_base import Base
 from nanobot.security.workspace_access import current_tool_workspace
+from nanobot.utils.cancellation import shield_and_drain
 from nanobot.utils.helpers import build_image_content_blocks, detect_image_mime
 
 
@@ -664,22 +667,31 @@ def _match_covers_line(match: _MatchSpan, line: int) -> bool:
     return match.line <= line <= _match_end_line(match)
 
 
-def _find_exact_matches(content: str, old_text: str) -> list[_MatchSpan]:
+def _find_exact_matches(
+    content: str,
+    old_text: str,
+    *,
+    max_matches: int | None = None,
+) -> list[_MatchSpan]:
     matches: list[_MatchSpan] = []
-    start = 0
+    search_start = 0
-    while True:
+    line_start = 0
-        idx = content.find(old_text, start)
+    line = 1
+    while max_matches is None or len(matches) < max_matches:
+        idx = content.find(old_text, search_start)
         if idx == -1:
             break
+        line += content.count("\n", line_start, idx)
         matches.append(
             _MatchSpan(
                 start=idx,
                 end=idx + len(old_text),
                 text=content[idx : idx + len(old_text)],
-                line=content.count("\n", 0, idx) + 1,
+                line=line,
             )
         )
-        start = idx + max(1, len(old_text))
+        line_start = idx
+        search_start = idx + max(1, len(old_text))
     return matches
 
 
@@ -735,27 +747,36 @@ def _find_quote_matches(content: str, old_text: str) -> list[_MatchSpan]:
     norm_content = _normalize_quotes(content)
     norm_old = _normalize_quotes(old_text)
     matches: list[_MatchSpan] = []
-    start = 0
+    search_start = 0
+    line_start = 0
+    line = 1
     while True:
-        idx = norm_content.find(norm_old, start)
+        idx = norm_content.find(norm_old, search_start)
         if idx == -1:
             break
+        line += content.count("\n", line_start, idx)
         matches.append(
             _MatchSpan(
                 start=idx,
                 end=idx + len(old_text),
                 text=content[idx : idx + len(old_text)],
-                line=content.count("\n", 0, idx) + 1,
+                line=line,
             )
         )
-        start = idx + max(1, len(norm_old))
+        line_start = idx
+        search_start = idx + max(1, len(norm_old))
     return matches
 
 
-def _find_matches(content: str, old_text: str) -> list[_MatchSpan]:
+def _find_matches(
-    """Locate all matches using progressively looser strategies."""
+    content: str,
+    old_text: str,
+    *,
+    max_exact_matches: int | None = None,
+) -> list[_MatchSpan]:
+    """Locate matches using progressively looser strategies."""
     for matcher in (
-        lambda: _find_exact_matches(content, old_text),
+        lambda: _find_exact_matches(content, old_text, max_matches=max_exact_matches),
         lambda: _find_trim_matches(content, old_text),
         lambda: _find_trim_matches(content, old_text, normalize_quotes=True),
         lambda: _find_quote_matches(content, old_text),
@@ -869,6 +890,43 @@ class EditFileTool(_FsTool):
         new_text: str | None = None,
         replace_all: bool = False, occurrence: int | None = None,
         line_hint: int | None = None, expected_replacements: int | None = None, **kwargs: Any,
+    ) -> str:
+        cancelled = threading.Event()
+        commit_lock = threading.Lock()
+        try:
+            return await asyncio.to_thread(
+                self._execute_sync,
+                path=path,
+                old_text=old_text,
+                new_text=new_text,
+                replace_all=replace_all,
+                occurrence=occurrence,
+                line_hint=line_hint,
+                expected_replacements=expected_replacements,
+                cancelled=cancelled,
+                commit_lock=commit_lock,
+            )
+        except asyncio.CancelledError:
+            cancelled.set()
+            # If a commit already started, do not report cancellation until the
+            # file bytes and FileStates record are settled. Otherwise, taking
+            # the lock first guarantees the worker observes ``cancelled`` before
+            # it can mutate the target.
+            await shield_and_drain(
+                asyncio.to_thread(self._wait_for_commit, commit_lock)
+            )
+            raise
+
+    @staticmethod
+    def _wait_for_commit(commit_lock: threading.Lock) -> None:
+        with commit_lock:
+            pass
+
+    def _execute_sync(
+        self, *, path: str | None, old_text: str | None,
+        new_text: str | None, replace_all: bool, occurrence: int | None,
+        line_hint: int | None, expected_replacements: int | None,
+        cancelled: threading.Event, commit_lock: threading.Lock,
     ) -> str:
         try:
             if not path:
@@ -892,6 +950,9 @@ class EditFileTool(_FsTool):
             # Create-file semantics: old_text='' + file doesn't exist → create
             if not file_exists:
                 if old_text == "":
+                    with commit_lock:
+                        if cancelled.is_set():
+                            return ToolResult.error("Error: edit_file cancelled.")
                         fp.parent.mkdir(parents=True, exist_ok=True)
                         fp.write_text(new_text, encoding="utf-8")
                         self._file_states.record_write(fp)
@@ -912,6 +973,9 @@ class EditFileTool(_FsTool):
                 content = raw.decode("utf-8")
                 if content.strip():
                     return ToolResult.error(f"Error: Cannot create file — {path} already exists and is not empty.")
+                with commit_lock:
+                    if cancelled.is_set():
+                        return ToolResult.error("Error: edit_file cancelled.")
                     fp.write_text(new_text, encoding="utf-8")
                     self._file_states.record_write(fp)
                 return f"Successfully edited {fp}"
@@ -923,7 +987,11 @@ class EditFileTool(_FsTool):
             uses_crlf = b"\r\n" in raw
             content = raw.decode("utf-8").replace("\r\n", "\n")
             norm_old = old_text.replace("\r\n", "\n")
-            matches = _find_matches(content, norm_old)
+            matches = _find_matches(
+                content,
+                norm_old,
+                max_exact_matches=occurrence,
+            )
 
             if not matches:
                 return self._not_found_msg(old_text, content, path)
@@ -1000,6 +1068,9 @@ class EditFileTool(_FsTool):
             if uses_crlf:
                 new_content = new_content.replace("\n", "\r\n")
 
+            with commit_lock:
+                if cancelled.is_set():
+                    return ToolResult.error("Error: edit_file cancelled.")
                 fp.write_bytes(new_content.encode("utf-8"))
                 self._file_states.record_write(fp)
             msg = f"Successfully edited {fp}"

nanobot/agent/tools/long_task.py

@@ -17,6 +17,7 @@ from nanobot.agent.tools.context import RequestContext, ToolContext, current_req
 from nanobot.agent.tools.schema import StringSchema, tool_parameters_schema
 from nanobot.bus.runtime_events import GoalStateChanged, RuntimeEventBus, RuntimeEventContext
 from nanobot.runtime_context import RuntimeContextBlock, wrap_runtime_context_lines
+from nanobot.session.async_compat import call_session_manager
 from nanobot.session.goal_state import (
     GOAL_STATE_KEY,
     MAX_GOAL_OBJECTIVE_CHARS,
@@ -28,6 +29,7 @@ from nanobot.session.goal_state import (
     sustained_goal_active,
 )
 from nanobot.session.turn_continuation import reset_goal_continuation_rounds
+from nanobot.utils.cancellation import shield_and_drain
 from nanobot.utils.prompt_templates import render_template
 
 if TYPE_CHECKING:
@@ -60,36 +62,68 @@ class _GoalToolsMixin:
         self._sessions = sessions
         self._runtime_events = runtime_events
 
-    def _session(self):
+    async def _get_or_create_session(self, key: str):
+        return await call_session_manager(
+            self._sessions,
+            "get_or_create_async",
+            self._sessions.get_or_create,
+            key,
+        )
+
+    async def _save_session(self, session: Any) -> None:
+        await call_session_manager(
+            self._sessions,
+            "save_async",
+            self._sessions.save,
+            session,
+        )
+
+    async def _session(self):
         request_ctx = current_request_context()
         if request_ctx is None:
             return None
         key = request_ctx.session_key
         if not key:
             return None
-        return self._sessions.get_or_create(key)
+        return await self._get_or_create_session(key)
 
     def _goal_mutation_allowed(self) -> bool:
         return current_request_context() is not None and goal_mutation_allowed()
 
-    def _save_goal_state(
+    async def _save_goal_state(
         self,
         sess: Any,
         blob: dict[str, Any],
         *,
         reset_continuation: bool = False,
+        revoke_permission: bool = False,
     ) -> None:
         previous_metadata = deepcopy(sess.metadata)
+        saved = False
+
+        async def save_and_publish() -> None:
+            nonlocal saved
             sess.metadata[GOAL_STATE_KEY] = blob
             discard_legacy_goal_state_key(sess.metadata)
             if reset_continuation:
                 reset_goal_continuation_rounds(sess.metadata)
             try:
-            self._sessions.save(sess)
+                await self._save_session(sess)
             except BaseException:
                 sess.metadata.clear()
                 sess.metadata.update(previous_metadata)
                 raise
+            saved = True
+            await self._publish_goal_state_changed(sess.metadata)
+
+        try:
+            await shield_and_drain(save_and_publish())
+        finally:
+            # This ContextVar belongs to the caller task, not the settlement task.
+            # Apply the post-save permission effect here even when cancellation was
+            # delayed until the durable save and runtime notification completed.
+            if revoke_permission and saved:
+                revoke_goal_mutation_permission()
 
     async def _publish_goal_state_changed(self, metadata: dict[str, Any]) -> None:
         runtime_events = self._runtime_events
@@ -175,7 +209,7 @@ class CreateGoalTool(Tool, _GoalToolsMixin):
     ) -> RuntimeContextBlock | None:
         if not request.session_key:
             return None
-        session = self._sessions.get_or_create(request.session_key)
+        session = await self._get_or_create_session(request.session_key)
         goal_start_requested = explicit_goal_requested(request.metadata)
         goal_active = sustained_goal_active(session.metadata)
         if not goal_start_requested and not goal_active:
@@ -197,7 +231,7 @@ class CreateGoalTool(Tool, _GoalToolsMixin):
         ui_summary: str | None = None,
         **kwargs: Any,
     ) -> str:
-        sess = self._session()
+        sess = await self._session()
         if sess is None:
             return ToolResult.error(
                 "Error: create_goal requires an active chat session (missing routing context)."
@@ -225,8 +259,7 @@ class CreateGoalTool(Tool, _GoalToolsMixin):
             "ui_summary": summary,
             "started_at": _iso_now(),
         }
-        self._save_goal_state(sess, blob, reset_continuation=True)
+        await self._save_goal_state(sess, blob, reset_continuation=True)
-        await self._publish_goal_state_changed(sess.metadata)
         extra = f"\nSummary line: {summary}" if summary else ""
         return (
             "Goal recorded. Keep working toward the objective using ordinary tools. "
@@ -305,7 +338,7 @@ class UpdateGoalTool(Tool, _GoalToolsMixin):
         ui_summary: str | None = None,
         **kwargs: Any,
     ) -> str:
-        sess = self._session()
+        sess = await self._session()
         if sess is None:
             return ToolResult.error("Error: update_goal requires an active chat session.")
         prior = parse_goal_state(goal_state_raw(sess.metadata))
@@ -340,8 +373,7 @@ class UpdateGoalTool(Tool, _GoalToolsMixin):
                 "previous_objective": str(prior.get("objective") or ""),
                 "recap": (recap or "").strip(),
             }
-            self._save_goal_state(sess, blob, reset_continuation=True)
+            await self._save_goal_state(sess, blob, reset_continuation=True)
-            await self._publish_goal_state_changed(sess.metadata)
             extra = f"\nSummary line: {summary}" if summary else ""
             return "Goal replaced. Continue toward the new objective using ordinary tools." + extra
 
@@ -359,9 +391,7 @@ class UpdateGoalTool(Tool, _GoalToolsMixin):
         }
         if normalized == "complete":
             blob["completed_at"] = ended
-        self._save_goal_state(sess, blob)
+        await self._save_goal_state(sess, blob, revoke_permission=True)
-        revoke_goal_mutation_permission()
-        await self._publish_goal_state_changed(sess.metadata)
 
         tail = (recap or "").strip()
         label = {

nanobot/agent/tools/mcp.py

@@ -20,10 +20,10 @@ from nanobot.agent.tools.base import Tool, ToolResult
 from nanobot.agent.tools.registry import ToolRegistry
 from nanobot.security.network import (
     PinnedDNSAsyncTransport,
+    async_resolve_url_target,
+    async_validate_url_target,
     env_proxy_applies_to_url,
     httpx_env_proxy_mounts,
-    resolve_url_target,
-    validate_url_target,
 )
 from nanobot.utils.cancellation import task_is_cancelling
 
@@ -249,7 +249,7 @@ async def _probe_http_url(url: str, timeout: float = 3.0) -> bool:
     port = parsed.port
     if not port:
         port = 443 if parsed.scheme == "https" else 80
-    ok, _, resolved_ips = resolve_url_target(url)
+    ok, _, resolved_ips = await async_resolve_url_target(url)
     if not ok:
         return False
     if env_proxy_applies_to_url(url):
@@ -298,7 +298,7 @@ def _pinned_transport_kwargs() -> dict[str, Any]:
 
 async def _validate_mcp_request_url(request: httpx.Request) -> None:
     """Validate each outgoing MCP HTTP request, including redirect targets."""
-    ok, error = validate_url_target(str(request.url))
+    ok, error = await async_validate_url_target(str(request.url))
     if not ok:
         raise httpx.RequestError(
             f"Blocked unsafe MCP URL {_redact_url(str(request.url))} ({error})",
@@ -1031,7 +1031,7 @@ async def connect_mcp_servers(
                     return False
 
             if transport_type in {"sse", "streamableHttp"}:
-                ok, error = validate_url_target(cfg.url)
+                ok, error = await async_validate_url_target(cfg.url)
                 if not ok:
                     logger.warning(
                         "MCP server '{}': blocked unsafe URL {} ({})",

nanobot/agent/tools/runtime_control.py

@@ -107,6 +107,13 @@ class RuntimeControl(Protocol):
         session_key: str | None,
     ) -> LLMRuntime: ...
 
+    async def set_model_preset_async(
+        self,
+        name: str,
+        *,
+        session_key: str | None,
+    ) -> LLMRuntime: ...
+
     def set_max_iterations(self, value: int) -> None: ...
 
     def set_context_window_tokens(self, value: int) -> LLMRuntime: ...
@@ -162,6 +169,12 @@ class _RuntimeControlTarget(Protocol):
 
     def set_session_model_preset(self, session_key: str, name: str) -> LLMRuntime: ...
 
+    async def set_session_model_preset_async(
+        self,
+        session_key: str,
+        name: str,
+    ) -> LLMRuntime: ...
+
 
 class AgentRuntimeControl:
     """Allowlisted adapter from agent-loop state to ``RuntimeControl``."""
@@ -208,6 +221,16 @@ class AgentRuntimeControl:
             return self.__target.set_session_model_preset(session_key, name)
         return self.__target.set_model_preset(name)
 
+    async def set_model_preset_async(
+        self,
+        name: str,
+        *,
+        session_key: str | None,
+    ) -> LLMRuntime:
+        if session_key is not None:
+            return await self.__target.set_session_model_preset_async(session_key, name)
+        return self.__target.set_model_preset(name)
+
     def set_max_iterations(self, value: int) -> None:
         self.__target.max_iterations = value
         self.__target.subagents.max_iterations = value

nanobot/agent/tools/search.py

@@ -4,9 +4,12 @@
 
 from __future__ import annotations
 
+import asyncio
 import fnmatch
 import os
 import re
+import threading
+import time
 from contextlib import suppress
 from pathlib import Path, PurePosixPath
 from typing import Any, Iterable, TypeVar
@@ -125,6 +128,8 @@ class _SearchTool(_FsTool):
 class FindFilesTool(_SearchTool):
     """Find files by path fragment, glob, or type."""
     _scopes = {"core", "subagent"}
+    _MAX_SCAN_PATHS = 500_000
+    _MAX_SCAN_SECONDS = 30.0
 
     @property
     def name(self) -> str:
@@ -218,7 +223,41 @@ class FindFilesTool(_SearchTool):
         offset: int = 0,
         **kwargs: Any,
     ) -> str:
+        cancelled = threading.Event()
         try:
+            return await asyncio.to_thread(
+                self._execute_sync,
+                path=path,
+                query=query,
+                glob=glob,
+                file_type=type,
+                include_dirs=include_dirs,
+                sort=sort,
+                head_limit=head_limit,
+                offset=offset,
+                cancelled=cancelled,
+            )
+        except asyncio.CancelledError:
+            cancelled.set()
+            raise
+        except PermissionError as e:
+            return ToolResult.error(f"Error: {e}")
+        except Exception as e:
+            return ToolResult.error(f"Error finding files: {e}")
+
+    def _execute_sync(
+        self,
+        *,
+        path: str,
+        query: str | None,
+        glob: str | None,
+        file_type: str | None,
+        include_dirs: bool,
+        sort: str,
+        head_limit: int | None,
+        offset: int,
+        cancelled: threading.Event,
+    ) -> str:
         target = self._resolve(path or ".")
         if not target.exists():
             return ToolResult.error(f"Error: Path not found: {path}")
@@ -235,8 +274,23 @@ class FindFilesTool(_SearchTool):
         )
         root = target if target.is_dir() else target.parent
         matches: list[tuple[str, float]] = []
+        deadline = time.monotonic() + self._MAX_SCAN_SECONDS
+        scanned = 0
 
         for candidate in self._iter_paths(target, include_dirs=include_dirs):
+            if cancelled.is_set():
+                raise RuntimeError("find_files scan cancelled")
+            scanned += 1
+            if scanned > self._MAX_SCAN_PATHS:
+                return ToolResult.error(
+                    f"Error: find_files scan exceeded {self._MAX_SCAN_PATHS} paths; "
+                    "narrow path, query, glob, or type and retry."
+                )
+            if time.monotonic() > deadline:
+                return ToolResult.error(
+                    f"Error: find_files scan exceeded {self._MAX_SCAN_SECONDS:g} seconds; "
+                    "narrow path, query, glob, or type and retry."
+                )
             if candidate.is_dir() and not include_dirs:
                 continue
             rel_path = candidate.relative_to(root).as_posix()
@@ -245,9 +299,9 @@ class FindFilesTool(_SearchTool):
 
             if glob and not _match_glob(rel_path, name, glob):
                 continue
-                if candidate.is_file() and not _matches_type(name, type):
+            if candidate.is_file() and not _matches_type(name, file_type):
                 continue
-                if candidate.is_dir() and type:
+            if candidate.is_dir() and file_type:
                 continue
             if not _matches_query(display_path, query):
                 continue
@@ -273,10 +327,6 @@ class FindFilesTool(_SearchTool):
         if note:
             result += "\n\n" + note
         return result
-        except PermissionError as e:
-            return ToolResult.error(f"Error: {e}")
-        except Exception as e:
-            return ToolResult.error(f"Error finding files: {e}")
 
 
 class GrepTool(_SearchTool):

nanobot/agent/tools/self.py

@@ -370,7 +370,7 @@ class MyTool(Tool):
         if not self._modify_allowed:
             return ToolResult.error("Error: set is disabled (tools.my.allow_set is false)")
         if action in ("modify", "set"):
-            return self._modify(key, value)
+            return await self._modify_async(key, value)
         return f"Unknown action: {action}"
 
     # -- inspect --
@@ -492,6 +492,11 @@ class MyTool(Tool):
             return ToolResult.error(f"Error: '{key}' is read-only and cannot be modified")
         return self._modify_scratchpad(key, value)
 
+    async def _modify_async(self, key: str | None, value: Any) -> str:
+        if key == "model_preset":
+            return await self._modify_model_preset_async(value)
+        return self._modify(key, value)
+
     def _modify_model_preset(self, value: Any) -> str:
         if not isinstance(value, str) or not value.strip():
             return ToolResult.error("Error: 'model_preset' must be a non-empty string")
@@ -520,6 +525,34 @@ class MyTool(Tool):
             f"context_window_tokens is now {runtime.context_window_tokens!r}"
         )
 
+    async def _modify_model_preset_async(self, value: Any) -> str:
+        if not isinstance(value, str) or not value.strip():
+            return ToolResult.error("Error: 'model_preset' must be a non-empty string")
+        name = value.strip()
+        session_key = current_request_session_key()
+        old = self._runtime_control.snapshot().model_preset
+        try:
+            runtime = await self._runtime_control.set_model_preset_async(
+                name,
+                session_key=session_key,
+            )
+        except (KeyError, ValueError) as exc:
+            message = str(exc.args[0]) if exc.args else str(exc)
+            punctuation = "" if message.endswith((".", "!", "?")) else "."
+            return ToolResult.error(f"Error: {message}{punctuation}")
+        if session_key:
+            self._audit("modify", f"model_preset = {name!r}")
+            return (
+                f"Set model_preset = {name!r} for the next turn; "
+                f"model will be {runtime.model!r}; "
+                f"context_window_tokens will be {runtime.context_window_tokens!r}"
+            )
+        self._audit("modify", f"model_preset: {old!r} -> {name!r}")
+        return (
+            f"Set model_preset = {name!r} (was {old!r}); model is now {runtime.model!r}; "
+            f"context_window_tokens is now {runtime.context_window_tokens!r}"
+        )
+
     def _modify_restricted(self, key: str, value: Any) -> str:
         spec = self.RESTRICTED[key]
         expected = cast(type[Any], spec["type"])

nanobot/agent/tools/shell.py

@@ -267,6 +267,7 @@ class ExecTool(Tool):
 
     _MAX_TIMEOUT = 600
     _MAX_OUTPUT = 10_000
+    _PREPARE_TIMEOUT_SECONDS = 6.0
 
     # Kernel device files safe as stdio redirect targets (#3599).
     _BENIGN_DEVICE_PATHS: frozenset[str] = frozenset({
@@ -324,7 +325,20 @@ class ExecTool(Tool):
         if max_output_chars is None:
             max_output_chars = max_output_tokens
 
-        prepared = self._prepare_command(command, working_dir, timeout, shell, login)
+        try:
+            prepared = await asyncio.wait_for(
+                asyncio.to_thread(
+                    self._prepare_command,
+                    command,
+                    working_dir,
+                    timeout,
+                    shell,
+                    login,
+                ),
+                timeout=self._PREPARE_TIMEOUT_SECONDS,
+            )
+        except asyncio.TimeoutError:
+            return ToolResult.error("Error: command validation timed out")
         if isinstance(prepared, str):
             return prepared
 
@@ -916,6 +930,15 @@ class ExecTool(Tool):
                     if self._is_benign_device_path(expanded):
                         continue
                 except Exception:
+                    # ``Path.expanduser()`` raises when a named user's home
+                    # cannot be resolved (notably on Windows).  An extracted
+                    # home path must fail closed rather than bypass the guard.
+                    if raw.strip().startswith("~"):
+                        return ToolResult.error(
+                            "Error: Command blocked by safety guard "
+                            "(path outside working dir)"
+                            + _WORKSPACE_BOUNDARY_NOTE
+                        )
                     continue
 
                 if self._is_benign_device_path(str(p)):

nanobot/agent/tools/web.py

@@ -96,7 +96,7 @@ def _normalize(text: str) -> str:
 
 
 def _validate_url(url: str) -> tuple[bool, str]:
-    """Validate URL scheme/domain. Does NOT check resolved IPs (use _validate_url_safe for that)."""
+    """Validate URL scheme/domain. Does not resolve IPs; use the async safe helper for that."""
     try:
         p = urlparse(url)
         if p.scheme not in ('http', 'https'):
@@ -108,18 +108,16 @@ def _validate_url(url: str) -> tuple[bool, str]:
         return False, str(e)
 
 
-def _validate_url_safe(url: str) -> tuple[bool, str]:
+async def _async_validate_url_safe(url: str) -> tuple[bool, str]:
-    """Validate URL with SSRF protection: scheme, domain, and resolved IP check."""
+    from nanobot.security.network import async_validate_url_target
-    from nanobot.security.network import validate_url_target
 
-    return validate_url_target(url)
+    return await async_validate_url_target(url)
 
 
-def _resolve_url_safe(url: str) -> tuple[bool, str, tuple[str, ...]]:
+async def _async_resolve_url_safe(url: str) -> tuple[bool, str, tuple[str, ...]]:
-    """Validate URL and return the resolved IPs to pin during the request."""
+    from nanobot.security.network import async_resolve_url_target
-    from nanobot.security.network import resolve_url_target
 
-    return resolve_url_target(url)
+    return await async_resolve_url_target(url)
 
 
 def _pinned_dns_transport() -> httpx.AsyncBaseTransport:
@@ -209,7 +207,7 @@ async def _get_with_safe_redirects(
     """GET a URL while validating every redirect target before requesting it."""
     current_url = url
     for _ in range(MAX_REDIRECTS + 1):
-        is_valid, error_msg, _ = _resolve_url_safe(current_url)
+        is_valid, error_msg, _ = await _async_resolve_url_safe(current_url)
         if not is_valid:
             return None, f"Redirect blocked: {error_msg}"
 
@@ -229,7 +227,7 @@ async def _get_with_safe_redirects(
             return response, None
 
         next_url = urljoin(str(response.url), location)
-        is_valid, error_msg = _validate_url_safe(next_url)
+        is_valid, error_msg = await _async_validate_url_safe(next_url)
         if not is_valid:
             await response.aclose()
             return None, f"Redirect blocked: {error_msg}"
@@ -249,7 +247,7 @@ async def _stream_with_safe_redirects(
     current_url = url
     chain_carries_credentials = _url_carries_credentials(url)
     for _ in range(MAX_REDIRECTS + 1):
-        is_valid, error_msg, _ = _resolve_url_safe(current_url)
+        is_valid, error_msg, _ = await _async_resolve_url_safe(current_url)
         if not is_valid:
             return None, None, f"Redirect blocked: {error_msg}", chain_carries_credentials
 
@@ -283,7 +281,7 @@ async def _stream_with_safe_redirects(
         chain_carries_credentials = (
             chain_carries_credentials or _url_carries_credentials(next_url)
         )
-        is_valid, error_msg = _validate_url_safe(next_url)
+        is_valid, error_msg = await _async_validate_url_safe(next_url)
         if not is_valid:
             await stream.__aexit__(None, None, None)
             return None, None, f"Redirect blocked: {error_msg}", chain_carries_credentials
@@ -1106,7 +1104,7 @@ class WebFetchTool(Tool):
         url = url.strip(" \t\r\n`\"'")
         extract_mode = kwargs.pop("extractMode", extract_mode)
         max_chars = cast(int, kwargs.pop("maxChars", max_chars) or self.max_chars)
-        is_valid, error_msg = _validate_url_safe(url)
+        is_valid, error_msg = await _async_validate_url_safe(url)
         if not is_valid:
             return json.dumps({"error": f"URL validation failed: {error_msg}", "url": url}, ensure_ascii=False)
 

nanobot/apps/cli/service.py

@@ -2,15 +2,20 @@
 
 from __future__ import annotations
 
+import asyncio
+import ctypes
 import json
 import os
 import re
 import shlex
 import shutil
+import signal
 import subprocess
 import sys
 import time
 from collections.abc import Iterable
+from contextlib import suppress
+from ctypes import wintypes
 from dataclasses import dataclass
 from importlib import metadata as importlib_metadata
 from pathlib import Path
@@ -97,6 +102,141 @@ class CliAppsRuntimeConfig:
     catalog_ttl_seconds: int = 3600
 
 
+@dataclass(slots=True)
+class _PreparedCliRun:
+    name: str
+    entry: str
+    resolved: str
+    args: list[str]
+    cwd: Path
+    timeout: int
+    env: dict[str, str]
+    artifact_snapshot: dict[Path, tuple[int, int]]
+
+
+class _JobObjectBasicLimitInformation(ctypes.Structure):
+    _fields_ = [
+        ("PerProcessUserTimeLimit", ctypes.c_int64),
+        ("PerJobUserTimeLimit", ctypes.c_int64),
+        ("LimitFlags", wintypes.DWORD),
+        ("MinimumWorkingSetSize", ctypes.c_size_t),
+        ("MaximumWorkingSetSize", ctypes.c_size_t),
+        ("ActiveProcessLimit", wintypes.DWORD),
+        ("Affinity", ctypes.c_size_t),
+        ("PriorityClass", wintypes.DWORD),
+        ("SchedulingClass", wintypes.DWORD),
+    ]
+
+
+class _IoCounters(ctypes.Structure):
+    _fields_ = [
+        ("ReadOperationCount", ctypes.c_uint64),
+        ("WriteOperationCount", ctypes.c_uint64),
+        ("OtherOperationCount", ctypes.c_uint64),
+        ("ReadTransferCount", ctypes.c_uint64),
+        ("WriteTransferCount", ctypes.c_uint64),
+        ("OtherTransferCount", ctypes.c_uint64),
+    ]
+
+
+class _JobObjectExtendedLimitInformation(ctypes.Structure):
+    _fields_ = [
+        ("BasicLimitInformation", _JobObjectBasicLimitInformation),
+        ("IoInfo", _IoCounters),
+        ("ProcessMemoryLimit", ctypes.c_size_t),
+        ("JobMemoryLimit", ctypes.c_size_t),
+        ("PeakProcessMemoryUsed", ctypes.c_size_t),
+        ("PeakJobMemoryUsed", ctypes.c_size_t),
+    ]
+
+
+class _WindowsJob:
+    """Best-effort Windows process tree ownership for timeout/cancellation."""
+
+    _KILL_ON_JOB_CLOSE = 0x00002000
+    _EXTENDED_LIMIT_INFORMATION = 9
+    _PROCESS_TERMINATE = 0x0001
+    _PROCESS_SET_QUOTA = 0x0100
+
+    def __init__(self) -> None:
+        win_dll = getattr(ctypes, "WinDLL")
+        self._kernel32 = win_dll("kernel32", use_last_error=True)
+        self._kernel32.CreateJobObjectW.argtypes = [wintypes.LPVOID, wintypes.LPCWSTR]
+        self._kernel32.CreateJobObjectW.restype = wintypes.HANDLE
+        self._kernel32.OpenProcess.argtypes = [wintypes.DWORD, wintypes.BOOL, wintypes.DWORD]
+        self._kernel32.OpenProcess.restype = wintypes.HANDLE
+        self._kernel32.AssignProcessToJobObject.argtypes = [wintypes.HANDLE, wintypes.HANDLE]
+        self._kernel32.AssignProcessToJobObject.restype = wintypes.BOOL
+        self._kernel32.SetInformationJobObject.argtypes = [
+            wintypes.HANDLE,
+            ctypes.c_int,
+            wintypes.LPVOID,
+            wintypes.DWORD,
+        ]
+        self._kernel32.SetInformationJobObject.restype = wintypes.BOOL
+        self._kernel32.TerminateJobObject.argtypes = [wintypes.HANDLE, wintypes.UINT]
+        self._kernel32.TerminateJobObject.restype = wintypes.BOOL
+        self._kernel32.CloseHandle.argtypes = [wintypes.HANDLE]
+        self._kernel32.CloseHandle.restype = wintypes.BOOL
+        self._handle: Any = self._kernel32.CreateJobObjectW(None, None)
+        if not self._handle:
+            raise OSError(ctypes.get_last_error(), "CreateJobObjectW failed")
+        try:
+            self._set_kill_on_close(True)
+        except OSError:
+            self._kernel32.CloseHandle(self._handle)
+            self._handle = None
+            raise
+
+    @classmethod
+    def create(cls) -> _WindowsJob | None:
+        if os.name != "nt":
+            return None
+        try:
+            return cls()
+        except OSError as exc:
+            logger.debug("CLI Apps: Windows job object unavailable: {}", exc)
+            return None
+
+    def _set_kill_on_close(self, enabled: bool) -> None:
+        info = _JobObjectExtendedLimitInformation()
+        info.BasicLimitInformation.LimitFlags = self._KILL_ON_JOB_CLOSE if enabled else 0
+        ok = self._kernel32.SetInformationJobObject(
+            self._handle,
+            self._EXTENDED_LIMIT_INFORMATION,
+            ctypes.byref(info),
+            ctypes.sizeof(info),
+        )
+        if not ok:
+            raise OSError(ctypes.get_last_error(), "SetInformationJobObject failed")
+
+    def assign(self, pid: int) -> bool:
+        process_handle = self._kernel32.OpenProcess(
+            self._PROCESS_TERMINATE | self._PROCESS_SET_QUOTA,
+            False,
+            pid,
+        )
+        if not process_handle:
+            return False
+        try:
+            return bool(self._kernel32.AssignProcessToJobObject(self._handle, process_handle))
+        finally:
+            self._kernel32.CloseHandle(process_handle)
+
+    def terminate(self) -> None:
+        if self._handle and not self._kernel32.TerminateJobObject(self._handle, 1):
+            raise OSError(ctypes.get_last_error(), "TerminateJobObject failed")
+
+    def close(self, *, kill_descendants: bool) -> None:
+        if not self._handle:
+            return
+        if not kill_descendants:
+            with suppress(OSError):
+                self._set_kill_on_close(False)
+        self._kernel32.CloseHandle(self._handle)
+        self._handle = None
+
+
 _BRANDS: dict[str, tuple[str, str]] = {
     "1password-cli": ("1password", "#3B66BC"),
     "arcgis": ("arcgis", "#2C7AC3"),
@@ -1428,6 +1568,197 @@ Use the `run_cli_app` tool with `name="{name}"` for command execution. Do not in
             lines.append(f"- {rel} ({kind}, {self._format_artifact_size(path)})")
         return lines
 
+    def _prepare_run(
+        self,
+        name: str,
+        args: list[str] | None,
+        *,
+        json_output: bool,
+        working_dir: str | None,
+        timeout: int | None,
+        restrict_to_workspace: bool,
+    ) -> _PreparedCliRun:
+        app = self.get_app(name)
+        installed = self._load_installed()
+        app_name = str(app["name"])
+        if app_name not in installed:
+            raise CliAppError(f"CLI app '{name}' is not installed")
+        cwd = self._resolve_cwd(working_dir, restrict_to_workspace=restrict_to_workspace)
+        entry = str(installed[app_name].get("entry_point") or app.get("entry_point") or "")
+        resolved = shutil.which(entry)
+        if not entry or not resolved:
+            raise CliAppError(f"{entry or name} is not available on PATH")
+        clean_args = [str(arg) for arg in (args or [])]
+        if json_output and "--json" not in clean_args:
+            clean_args = ["--json", *clean_args]
+        effective_timeout = max(1, min(timeout or self.runtime.run_timeout, 600))
+        return _PreparedCliRun(
+            name=name,
+            entry=entry,
+            resolved=resolved,
+            args=clean_args,
+            cwd=cwd,
+            timeout=effective_timeout,
+            env=self._subprocess_env(),
+            artifact_snapshot=self._artifact_snapshot(cwd),
+        )
+
+    def _format_run_result(
+        self,
+        prepared: _PreparedCliRun,
+        *,
+        returncode: int,
+        stdout: str,
+        stderr: str,
+    ) -> str:
+        command = " ".join([prepared.entry, *(shlex.quote(arg) for arg in prepared.args)])
+        output = [
+            f"CLI app '{prepared.name}' exited {returncode}.",
+            f"Command: {command}",
+        ]
+        if stdout:
+            output.append("\nSTDOUT:\n" + stdout.rstrip())
+        if stderr:
+            output.append("\nSTDERR:\n" + stderr.rstrip())
+        artifacts = self._changed_artifacts(prepared.cwd, prepared.artifact_snapshot)
+        if artifacts:
+            output.append(
+                "\nArtifacts created or updated:\n"
+                + "\n".join(self._format_artifact_lines(prepared.cwd, artifacts))
+            )
+            if any(path.suffix.lower() in _INLINE_ARTIFACT_EXTENSIONS for path in artifacts):
+                output.append(
+                    "\nTo show a preview in WebUI, reference a raster artifact with Markdown "
+                    "using its workspace-relative path, for example `![diagram](diagram.png)`."
+                )
+        return _truncate("\n".join(output))
+
+    @staticmethod
+    def _terminate_run_process_sync(
+        process: subprocess.Popen[str],
+        job: _WindowsJob | None,
+    ) -> None:
+        if job is not None:
+            with suppress(OSError):
+                job.terminate()
+            job.close(kill_descendants=True)
+        elif os.name == "nt":
+            with suppress(OSError, subprocess.TimeoutExpired):
+                subprocess.run(
+                    ["taskkill", "/PID", str(process.pid), "/T", "/F"],
+                    check=False,
+                    stdout=subprocess.DEVNULL,
+                    stderr=subprocess.DEVNULL,
+                    timeout=5,
+                )
+        else:
+            with suppress(ProcessLookupError, PermissionError):
+                os.killpg(process.pid, signal.SIGKILL)
+        if process.poll() is None:
+            with suppress(ProcessLookupError):
+                process.kill()
+        with suppress(subprocess.TimeoutExpired):
+            process.wait(timeout=5)
+
+    @staticmethod
+    async def _terminate_run_process(
+        process: asyncio.subprocess.Process,
+        job: _WindowsJob | None,
+    ) -> None:
+        if job is not None:
+            with suppress(OSError):
+                await asyncio.to_thread(job.terminate)
+            job.close(kill_descendants=True)
+        elif os.name == "nt":
+            with suppress(OSError, asyncio.TimeoutError):
+                await asyncio.wait_for(
+                    asyncio.to_thread(
+                        subprocess.run,
+                        ["taskkill", "/PID", str(process.pid), "/T", "/F"],
+                        check=False,
+                        stdout=subprocess.DEVNULL,
+                        stderr=subprocess.DEVNULL,
+                        timeout=5,
+                    ),
+                    timeout=6.0,
+                )
+        else:
+            with suppress(ProcessLookupError, PermissionError):
+                os.killpg(process.pid, signal.SIGKILL)
+        if process.returncode is None:
+            with suppress(ProcessLookupError):
+                process.kill()
+        with suppress(asyncio.TimeoutError, ProcessLookupError):
+            await asyncio.wait_for(process.wait(), timeout=5.0)
+
+    async def run_async(
+        self,
+        name: str,
+        args: list[str] | None = None,
+        *,
+        json_output: bool = False,
+        working_dir: str | None = None,
+        timeout: int | None = None,
+        restrict_to_workspace: bool = False,
+    ) -> str:
+        prepared = await asyncio.to_thread(
+            self._prepare_run,
+            name,
+            args,
+            json_output=json_output,
+            working_dir=working_dir,
+            timeout=timeout,
+            restrict_to_workspace=restrict_to_workspace,
+        )
+        process_kwargs: dict[str, Any] = {}
+        if os.name == "nt":
+            process_kwargs["creationflags"] = subprocess.CREATE_NEW_PROCESS_GROUP
+        else:
+            process_kwargs["start_new_session"] = True
+        job = _WindowsJob.create()
+        try:
+            process = await asyncio.create_subprocess_exec(
+                prepared.resolved,
+                *prepared.args,
+                cwd=str(prepared.cwd),
+                stdout=asyncio.subprocess.PIPE,
+                stderr=asyncio.subprocess.PIPE,
+                env=prepared.env,
+                **process_kwargs,
+            )
+        except BaseException:
+            if job is not None:
+                job.close(kill_descendants=False)
+            raise
+        if job is not None and not job.assign(process.pid):
+            job.close(kill_descendants=False)
+            job = None
+        try:
+            stdout_raw, stderr_raw = await asyncio.wait_for(
+                process.communicate(),
+                timeout=prepared.timeout,
+            )
+        except asyncio.TimeoutError:
+            await self._terminate_run_process(process, job)
+            return f"CLI app '{prepared.name}' timed out after {prepared.timeout}s"
+        except asyncio.CancelledError:
+            await self._terminate_run_process(process, job)
+            raise
+        except BaseException:
+            await self._terminate_run_process(process, job)
+            raise
+        if job is not None:
+            job.close(kill_descendants=False)
+        stdout = stdout_raw.decode("utf-8", errors="replace")
+        stderr = stderr_raw.decode("utf-8", errors="replace")
+        return await asyncio.to_thread(
+            self._format_run_result,
+            prepared,
+            returncode=process.returncode or 0,
+            stdout=stdout,
+            stderr=stderr,
+        )
+
     def run(
         self,
         name: str,
@@ -1438,50 +1769,52 @@ Use the `run_cli_app` tool with `name="{name}"` for command execution. Do not in
         timeout: int | None = None,
         restrict_to_workspace: bool = False,
     ) -> str:
-        app = self.get_app(name)
+        prepared = self._prepare_run(
-        installed = self._load_installed()
+            name,
-        if str(app["name"]) not in installed:
+            args,
-            raise CliAppError(f"CLI app '{name}' is not installed")
+            json_output=json_output,
-        cwd = self._resolve_cwd(working_dir, restrict_to_workspace=restrict_to_workspace)
+            working_dir=working_dir,
-        entry = str(installed[str(app["name"])].get("entry_point") or app.get("entry_point") or "")
+            timeout=timeout,
-        resolved = shutil.which(entry)
+            restrict_to_workspace=restrict_to_workspace,
-        if not entry or not resolved:
+        )
-            raise CliAppError(f"{entry or name} is not available on PATH")
+        process_kwargs: dict[str, Any] = {}
-        clean_args = [str(arg) for arg in (args or [])]
+        if os.name == "nt":
-        if json_output and "--json" not in clean_args:
+            process_kwargs["creationflags"] = subprocess.CREATE_NEW_PROCESS_GROUP
-            clean_args = ["--json", *clean_args]
+        else:
-        effective_timeout = max(1, min(timeout or self.runtime.run_timeout, 600))
+            process_kwargs["start_new_session"] = True
-        artifact_snapshot = self._artifact_snapshot(cwd)
+        job = _WindowsJob.create()
         try:
-            result = subprocess.run(
+            process = subprocess.Popen(
-                [resolved, *clean_args],
+                [prepared.resolved, *prepared.args],
-                cwd=str(cwd),
+                cwd=str(prepared.cwd),
-                capture_output=True,
+                stdout=subprocess.PIPE,
+                stderr=subprocess.PIPE,
                 text=True,
                 encoding="utf-8",
                 errors="replace",
-                timeout=effective_timeout,
+                env=prepared.env,
-                env=self._subprocess_env(),
+                **process_kwargs,
             )
+        except BaseException:
+            if job is not None:
+                job.close(kill_descendants=False)
+            raise
+        if job is not None and not job.assign(process.pid):
+            job.close(kill_descendants=False)
+            job = None
+        try:
+            stdout, stderr = process.communicate(timeout=prepared.timeout)
         except subprocess.TimeoutExpired:
-            return f"CLI app '{name}' timed out after {effective_timeout}s"
+            self._terminate_run_process_sync(process, job)
-        output = [
+            return f"CLI app '{prepared.name}' timed out after {prepared.timeout}s"
-            f"CLI app '{name}' exited {result.returncode}.",
+        except BaseException:
-            f"Command: {entry} {' '.join(shlex.quote(arg) for arg in clean_args)}".rstrip(),
+            self._terminate_run_process_sync(process, job)
-        ]
+            raise
-        if result.stdout:
+        if job is not None:
-            output.append("\nSTDOUT:\n" + result.stdout.rstrip())
+            job.close(kill_descendants=False)
-        if result.stderr:
+        return self._format_run_result(
-            output.append("\nSTDERR:\n" + result.stderr.rstrip())
+            prepared,
-        artifacts = self._changed_artifacts(cwd, artifact_snapshot)
+            returncode=process.returncode,
-        if artifacts:
+            stdout=stdout,
-            output.append(
+            stderr=stderr,
-                "\nArtifacts created or updated:\n"
-                + "\n".join(self._format_artifact_lines(cwd, artifacts))
         )
-            if any(path.suffix.lower() in _INLINE_ARTIFACT_EXTENSIONS for path in artifacts):
-                output.append(
-                    "\nTo show a preview in WebUI, reference a raster artifact with Markdown "
-                    "using its workspace-relative path, for example `![diagram](diagram.png)`."
-                )
-        return _truncate("\n".join(output))

nanobot/channels/dingtalk/runtime.py

@@ -21,7 +21,10 @@ from nanobot.bus.events import OutboundMessage
 from nanobot.bus.queue import MessageBus
 from nanobot.channels.base import BaseChannel
 from nanobot.config.schema import Base
-from nanobot.security.network import validate_resolved_url, validate_url_target
+from nanobot.security.network import (
+    async_validate_resolved_url,
+    async_validate_url_target,
+)
 
 DINGTALK_MAX_REMOTE_MEDIA_BYTES = 20 * 1024 * 1024
 DINGTALK_MAX_REMOTE_MEDIA_REDIRECTS = 3
@@ -417,8 +420,8 @@ class DingTalkChannel(BaseChannel):
             return self._zip_bytes(filename, data)
         return data, filename, content_type
 
-    def _validate_remote_media_url(self, media_ref: str) -> bool:
+    async def _validate_remote_media_url(self, media_ref: str) -> bool:
-        ok, err = validate_url_target(media_ref)
+        ok, err = await async_validate_url_target(media_ref)
         if not ok:
             self.logger.warning("remote media URL blocked ref={} reason={}", media_ref, err)
             return False
@@ -434,7 +437,11 @@ class DingTalkChannel(BaseChannel):
         allowed_hosts = {host.lower() for host in self.config.remote_media_redirect_allowed_hosts}
         return next_host in allowed_hosts
 
-    def _next_remote_media_url(self, current_url: str, location: str | None) -> str | None:
+    async def _next_remote_media_url(
+        self,
+        current_url: str,
+        location: str | None,
+    ) -> str | None:
         if not self.config.allow_remote_media_redirects:
             self.logger.warning("media download redirect refused ref={}", current_url)
             return None
@@ -449,7 +456,7 @@ class DingTalkChannel(BaseChannel):
                 next_url,
             )
             return None
-        if not self._validate_remote_media_url(next_url):
+        if not await self._validate_remote_media_url(next_url):
             return None
         return next_url
 
@@ -461,7 +468,7 @@ class DingTalkChannel(BaseChannel):
         if not self._http:
             return None, None
 
-        if not self._validate_remote_media_url(media_ref):
+        if not await self._validate_remote_media_url(media_ref):
             return None, None
 
         try:
@@ -473,7 +480,7 @@ class DingTalkChannel(BaseChannel):
                 current_url = media_ref
                 for _ in range(DINGTALK_MAX_REMOTE_MEDIA_REDIRECTS + 1):
                     async with stream("GET", current_url, follow_redirects=False) as resp:
-                        final_ok, final_err = validate_resolved_url(str(resp.url))
+                        final_ok, final_err = await async_validate_resolved_url(str(resp.url))
                         if not final_ok:
                             self.logger.warning(
                                 "remote media redirect blocked ref={} final={} reason={}",
@@ -483,7 +490,7 @@ class DingTalkChannel(BaseChannel):
                             )
                             return None, None
                         if 300 <= resp.status_code < 400:
-                            next_url = self._next_remote_media_url(
+                            next_url = await self._next_remote_media_url(
                                 str(resp.url), resp.headers.get("location")
                             )
                             if not next_url:
@@ -516,7 +523,9 @@ class DingTalkChannel(BaseChannel):
             current_url = media_ref
             for _ in range(DINGTALK_MAX_REMOTE_MEDIA_REDIRECTS + 1):
                 resp = await self._http.get(current_url, follow_redirects=False)
-                final_ok, final_err = validate_resolved_url(str(getattr(resp, "url", current_url)))
+                final_ok, final_err = await async_validate_resolved_url(
+                    str(getattr(resp, "url", current_url))
+                )
                 if not final_ok:
                     self.logger.warning(
                         "remote media redirect blocked ref={} final={} reason={}",
@@ -526,7 +535,7 @@ class DingTalkChannel(BaseChannel):
                     )
                     return None, None
                 if 300 <= resp.status_code < 400:
-                    next_url = self._next_remote_media_url(
+                    next_url = await self._next_remote_media_url(
                         str(getattr(resp, "url", current_url)), resp.headers.get("location")
                     )
                     if not next_url:

nanobot/channels/napcat/runtime.py

@@ -24,7 +24,7 @@ from nanobot.bus.queue import MessageBus
 from nanobot.channels.base import BaseChannel
 from nanobot.config.paths import get_media_dir
 from nanobot.config.schema import Base
-from nanobot.security.network import validate_url_target
+from nanobot.security.network import async_validate_url_target
 from nanobot.utils.helpers import safe_filename
 
 _DOWNLOAD_TIMEOUT = aiohttp.ClientTimeout(total=60)
@@ -473,7 +473,7 @@ class NapcatChannel(BaseChannel):
         if not ref:
             return None
         if ref.startswith(("http://", "https://")):
-            ok, err = validate_url_target(ref)
+            ok, err = await async_validate_url_target(ref)
             if not ok:
                 logger.warning("napcat: rejected remote image '{}': {}", ref, err)
                 return None
@@ -525,7 +525,7 @@ class NapcatChannel(BaseChannel):
         # logger.debug("napcat: downloading image from {}", url)
         if self._http is None:
             return None
-        ok, err = validate_url_target(url)
+        ok, err = await async_validate_url_target(url)
         if not ok:
             logger.warning("napcat: skip image '{}': {}", url, err)
             return None

nanobot/channels/napcat/tests/test_napcat_channel.py

@@ -149,9 +149,13 @@ async def test_download_image_rejects_redirects(tmp_path, monkeypatch) -> None:
     channel = _channel()
     channel._media_root = tmp_path
     channel._http = _FakeHttp(_FakeResponse(status=302))
+
+    async def allow_url(_url: str) -> tuple[bool, str]:
+        return True, ""
+
     monkeypatch.setattr(
-        "nanobot.channels.napcat.runtime.validate_url_target",
+        "nanobot.channels.napcat.runtime.async_validate_url_target",
-        lambda _url: (True, ""),
+        allow_url,
     )
 
     result = await channel._download_image({"url": "https://example.com/a.png", "file": "a.png"})

nanobot/channels/qq/runtime.py

@@ -40,7 +40,7 @@ from nanobot.bus.events import OutboundMessage
 from nanobot.bus.queue import MessageBus
 from nanobot.channels.base import BaseChannel
 from nanobot.config.schema import Base
-from nanobot.security.network import validate_url_target
+from nanobot.security.network import async_validate_url_target
 from nanobot.utils.logging_bridge import redirect_lib_logging
 
 try:
@@ -458,7 +458,7 @@ class QQChannel(BaseChannel):
                 return None, None
 
         # Remote URL
-        ok, err = validate_url_target(media_ref)
+        ok, err = await async_validate_url_target(media_ref)
         if not ok:
             self.logger.warning("outbound media URL validation failed url={} err={}", media_ref, err)
             return None, None

nanobot/channels/slack/runtime.py

@@ -23,8 +23,8 @@ from nanobot.config.schema import Base
 from nanobot.pairing import is_approved
 from nanobot.security.network import (
     PinnedDNSAsyncTransport,
+    async_validate_url_target,
     httpx_env_proxy_mounts,
-    validate_url_target,
 )
 from nanobot.utils.helpers import safe_filename, split_message
 
@@ -95,7 +95,7 @@ _HTML_DOWNLOAD_PREFIXES = (b"<!doctype html", b"<html")
 
 async def _validate_slack_download_request(request: httpx.Request) -> None:
     """Validate every Slack file request, including redirects, before transport."""
-    ok, error = validate_url_target(str(request.url))
+    ok, error = await async_validate_url_target(str(request.url))
     if not ok:
         raise httpx.RequestError(f"unsafe Slack file URL: {error}", request=request)
 

nanobot/channels/slack/tests/test_slack_channel.py

@@ -859,13 +859,13 @@ def _patch_download_validation(
     monkeypatch: pytest.MonkeyPatch,
     validated: list[str],
 ) -> None:
-    def validate(url: str) -> tuple[bool, str]:
+    async def validate(url: str) -> tuple[bool, str]:
         validated.append(url)
         if "169.254.169.254" in url:
             return False, "blocked metadata address"
         return True, ""
 
-    monkeypatch.setattr("nanobot.channels.slack.runtime.validate_url_target", validate)
+    monkeypatch.setattr("nanobot.channels.slack.runtime.async_validate_url_target", validate)
 
 
 @pytest.mark.asyncio

nanobot/channels/telegram/runtime.py

@@ -36,7 +36,7 @@ from nanobot.channels.base import BaseChannel
 from nanobot.command.builtin import build_help_text
 from nanobot.config.paths import get_media_dir
 from nanobot.config.schema import Base
-from nanobot.security.network import validate_url_target
+from nanobot.security.network import async_validate_url_target
 from nanobot.utils.helpers import split_message
 from nanobot.utils.logging_bridge import redirect_lib_logging
 
@@ -956,7 +956,7 @@ class TelegramChannel(BaseChannel):
 
                 # Telegram Bot API accepts HTTP(S) URLs directly for media params.
                 if self._is_remote_media_url(media_path):
-                    ok, error = validate_url_target(media_path)
+                    ok, error = await async_validate_url_target(media_path)
                     if not ok:
                         raise ValueError(f"unsafe media URL: {error}")
                     await self._call_with_retry(

nanobot/channels/telegram/tests/test_telegram_channel.py

@@ -1488,7 +1488,14 @@ async def test_send_remote_media_url_after_security_validation(monkeypatch) -> N
         MessageBus(),
     )
     _install_ready_app(channel)
-    monkeypatch.setattr("nanobot.channels.telegram.runtime.validate_url_target", lambda url: (True, ""))
+
+    async def allow_url(_url: str) -> tuple[bool, str]:
+        return True, ""
+
+    monkeypatch.setattr(
+        "nanobot.channels.telegram.runtime.async_validate_url_target",
+        allow_url,
+    )
 
     await channel.send(
         OutboundMessage(
@@ -1546,9 +1553,13 @@ async def test_send_blocks_unsafe_remote_media_url(monkeypatch) -> None:
         MessageBus(),
     )
     _install_ready_app(channel)
+
+    async def deny_url(_url: str) -> tuple[bool, str]:
+        return False, "Blocked: example.com resolves to private/internal address 127.0.0.1"
+
     monkeypatch.setattr(
-        "nanobot.channels.telegram.runtime.validate_url_target",
+        "nanobot.channels.telegram.runtime.async_validate_url_target",
-        lambda url: (False, "Blocked: example.com resolves to private/internal address 127.0.0.1"),
+        deny_url,
     )
 
     await channel.send(

nanobot/channels/websocket/runtime.py

@@ -55,6 +55,7 @@ from nanobot.security.workspace_access import (
     WORKSPACE_SCOPE_METADATA_KEY,
     WorkspaceScopeError,
 )
+from nanobot.session.async_compat import call_session_manager
 from nanobot.session.goal_state import goal_state_ws_blob
 from nanobot.session.model_selection import model_preset_from_metadata
 from nanobot.session.recovery import recovery_state_from_metadata
@@ -447,6 +448,26 @@ class WebSocketChannel(BaseChannel):
         if sessions is None:
             return {}
         snapshot = sessions.read_session_metadata(f"websocket:{chat_id}")
+        return self._attached_model_fields_from_snapshot(chat_id, snapshot)
+
+    async def _attached_model_fields_async(self, chat_id: str) -> dict[str, Any]:
+        """Build attach fields without blocking the gateway event loop."""
+        sessions = self.gateway.session_manager
+        if sessions is None:
+            return {}
+        snapshot = await call_session_manager(
+            sessions,
+            "read_session_metadata_async",
+            sessions.read_session_metadata,
+            f"websocket:{chat_id}",
+        )
+        return self._attached_model_fields_from_snapshot(chat_id, snapshot)
+
+    def _attached_model_fields_from_snapshot(
+        self,
+        chat_id: str,
+        snapshot: dict[str, Any] | None,
+    ) -> dict[str, Any]:
         raw_metadata = snapshot.get("metadata") if snapshot is not None else None
         metadata = cast(dict[str, object], raw_metadata) if isinstance(raw_metadata, dict) else None
         fields: dict[str, Any] = {}
@@ -510,13 +531,16 @@ class WebSocketChannel(BaseChannel):
         fork_key: str,
     ) -> None:
         """Attach and hydrate a newly created WebUI chat fork."""
-        scope = self._workspaces.scope_for_session_key(fork_key)
+        scope = await asyncio.to_thread(
+            self._workspaces.scope_for_session_key,
+            fork_key,
+        )
         self._attach(connection, fork_id)
         await self._send_event(
             connection,
             "attached",
             chat_id=fork_id,
-            **self._attached_model_fields(fork_id),
+            **await self._attached_model_fields_async(fork_id),
         )
         await self._send_event(
             connection,
@@ -904,7 +928,7 @@ class WebSocketChannel(BaseChannel):
                 connection,
                 "attached",
                 chat_id=new_id,
-                **self._attached_model_fields(new_id),
+                **await self._attached_model_fields_async(new_id),
             )
             await self._send_event(
                 connection,
@@ -960,7 +984,7 @@ class WebSocketChannel(BaseChannel):
                 connection,
                 "attached",
                 chat_id=cid,
-                **self._attached_model_fields(cid),
+                **await self._attached_model_fields_async(cid),
             )
             await self._hydrate_after_subscribe(cid)
             return
@@ -1263,7 +1287,7 @@ class WebSocketChannel(BaseChannel):
                         else False
                     ),
                 )
-                self._workspaces.persist_scope(cid, scope)
+                await asyncio.to_thread(self._workspaces.persist_scope, cid, scope)
                 accepted = True
             finally:
                 if not accepted and queued_owner is not None:
@@ -1553,7 +1577,7 @@ class WebSocketChannel(BaseChannel):
         turn_id: str | None = None,
     ) -> Any | None:
         try:
-            return resolver()
+            return await asyncio.to_thread(resolver)
         except WorkspaceScopeError as exc:
             await self._send_event(
                 connection,

nanobot/channels/websocket/tests/test_websocket_http_routes.py

@@ -4,6 +4,7 @@ import asyncio
 import json
 import random
 import socket
+import threading
 import time
 from contextlib import suppress
 from pathlib import Path
@@ -2576,6 +2577,69 @@ async def test_webui_automations_route_lists_all_jobs_and_allows_user_actions(
         await server_task
 
 
+@pytest.mark.asyncio
+async def test_webui_cron_update_rearms_started_service_on_owner_loop(
+    bus: MagicMock,
+    tmp_path: Path,
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    store_path = tmp_path / "cron" / "jobs.json"
+    cron = CronService(store_path, max_sleep_ms=60_000)
+    job = cron.add_job(
+        name="Before update",
+        schedule=CronSchedule(kind="every", every_ms=86_400_000),
+        message="Check the repo status",
+        session_key="websocket:abc",
+        origin_channel="websocket",
+        origin_chat_id="abc",
+    )
+    await cron.start()
+
+    owner_thread_id = threading.get_ident()
+    initial_timer = cron._timer_task
+    request_thread_ids: list[int] = []
+    arm_thread_ids: list[int] = []
+    timer_rearmed = asyncio.Event()
+    original_request_timer_rearm = cron._request_timer_rearm
+    original_arm_timer = cron._arm_timer
+
+    def tracked_request_timer_rearm() -> None:
+        request_thread_ids.append(threading.get_ident())
+        original_request_timer_rearm()
+
+    def tracked_arm_timer() -> None:
+        arm_thread_ids.append(threading.get_ident())
+        original_arm_timer()
+        timer_rearmed.set()
+
+    monkeypatch.setattr(cron, "_request_timer_rearm", tracked_request_timer_rearm)
+    monkeypatch.setattr(cron, "_arm_timer", tracked_arm_timer)
+    channel = _ch(bus, cron_service=cron, port=_free_port())
+
+    try:
+        response = await _webui_mutate(
+            channel,
+            "automation.update",
+            {"id": job.id, "values": {"name": "After update"}},
+        )
+        await asyncio.wait_for(timer_rearmed.wait(), timeout=1)
+
+        assert response.status_code == 200
+        assert request_thread_ids
+        assert all(thread_id != owner_thread_id for thread_id in request_thread_ids)
+        assert arm_thread_ids and set(arm_thread_ids) == {owner_thread_id}
+        assert cron._timer_task is not None
+        assert cron._timer_task is not initial_timer
+        assert not cron._timer_task.done()
+
+        stored = json.loads(store_path.read_text(encoding="utf-8"))
+        assert len(stored["jobs"]) == 1
+        assert stored["jobs"][0]["name"] == "After update"
+    finally:
+        cron.stop()
+        await asyncio.sleep(0)
+
+
 @pytest.mark.asyncio
 async def test_webui_automations_route_manages_local_triggers(
     bus: MagicMock, tmp_path: Path
@@ -3769,3 +3833,77 @@ def test_bootstrap_secret_also_enforced_on_localhost(bus: MagicMock) -> None:
     channel = _ch(bus, host="0.0.0.0", tokenIssueSecret="s3cret")
     resp = channel.gateway.http._handle_bootstrap(_LOCAL, _NO_HEADERS)
     assert resp.status_code == 401
+
+
+@pytest.mark.asyncio
+async def test_webui_skill_update_cancellation_waits_for_config_and_runtime_state(
+    bus: MagicMock,
+    tmp_path: Path,
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    from nanobot.webui import ws_http
+
+    skill_dir = tmp_path / "skills" / "cancel-safe-skill"
+    skill_dir.mkdir(parents=True)
+    (skill_dir / "SKILL.md").write_text(
+        "---\nname: cancel-safe-skill\ndescription: Cancellation test skill.\n---\n",
+        encoding="utf-8",
+    )
+    mutation_started = threading.Event()
+    release_mutation = threading.Event()
+    original_update = ws_http.set_webui_skill_enabled
+    update_calls = 0
+
+    def blocked_update(*args: Any, **kwargs: Any) -> dict[str, Any]:
+        nonlocal update_calls
+        update_calls += 1
+        mutation_started.set()
+        assert release_mutation.wait(timeout=1)
+        return original_update(*args, **kwargs)
+
+    monkeypatch.setattr(ws_http, "set_webui_skill_enabled", blocked_update)
+    channel = _ch(
+        bus,
+        session_manager=_seed_session(tmp_path),
+        workspace_path=tmp_path,
+        port=_free_port(),
+    )
+    runtime_states: list[set[str]] = []
+    channel.gateway.http.skill_state_action = runtime_states.append
+    task = asyncio.create_task(
+        _webui_mutate(
+            channel,
+            "skill.update",
+            {"name": "cancel-safe-skill", "enabled": False},
+        )
+    )
+
+    assert await asyncio.to_thread(mutation_started.wait, 1)
+    try:
+        task.cancel()
+        await asyncio.sleep(0)
+        assert not task.done()
+        assert runtime_states == []
+    finally:
+        release_mutation.set()
+
+    with pytest.raises(asyncio.CancelledError):
+        await asyncio.wait_for(task, timeout=1)
+
+    assert update_calls == 1
+    assert "cancel-safe-skill" in channel.gateway.http.disabled_skills
+    assert runtime_states == [{"cancel-safe-skill"}]
+    saved = load_config(channel.gateway.settings.config.path)
+    assert "cancel-safe-skill" in saved.agents.defaults.disabled_skills
+
+    settled_state = (
+        update_calls,
+        set(channel.gateway.http.disabled_skills),
+        list(runtime_states),
+    )
+    await asyncio.sleep(0.05)
+    assert (
+        update_calls,
+        set(channel.gateway.http.disabled_skills),
+        runtime_states,
+    ) == settled_state

nanobot/cli/gateway_runtime.py

@@ -34,6 +34,7 @@ from nanobot.config.paths import is_default_workspace
 from nanobot.config.schema import Config
 from nanobot.gateway.runtime import GatewayInstance
 from nanobot.security.network import is_loopback_host
+from nanobot.session.async_compat import call_session_manager as _call_session_manager
 from nanobot.session.keys import UNIFIED_SESSION_KEY, last_channel_from_metadata
 from nanobot.utils.evaluator import evaluate_response, resolve_evaluator_prompt
 from nanobot.utils.helpers import sync_workspace_templates
@@ -45,6 +46,30 @@ __all__ = ["_run_gateway"]
 
 console = Console()
 
+_EVENT_LOOP_LAG_INTERVAL_S = 0.5
+_EVENT_LOOP_LAG_WARNING_S = 0.25
+
+
+async def _monitor_event_loop_lag(
+    *,
+    interval_s: float = _EVENT_LOOP_LAG_INTERVAL_S,
+    warning_threshold_s: float = _EVENT_LOOP_LAG_WARNING_S,
+    log: Any | None = None,
+) -> None:
+    """Log scheduler drift so gateway-wide stalls have direct evidence."""
+    loop = asyncio.get_running_loop()
+    lag_log = log or logger
+    while True:
+        expected = loop.time() + interval_s
+        await asyncio.sleep(interval_s)
+        lag_s = max(0.0, loop.time() - expected)
+        if lag_s >= warning_threshold_s:
+            lag_log.warning(
+                "event loop lag operation=gateway duration_ms={} interval_ms={}",
+                int(lag_s * 1000),
+                int(interval_s * 1000),
+            )
+
 
 def _http_endpoint_responding(url: str, *, timeout_s: float = 0.25) -> bool:
     """Return whether an HTTP endpoint responds, including with an auth error."""
@@ -493,12 +518,22 @@ def _run_gateway(
             and hasattr(session_manager, "save")
         ):
             key = session_key or _channel_session_key(msg.channel, msg.chat_id)
-            session = session_manager.get_or_create(key)
+            session = await _call_session_manager(
+                session_manager,
+                "get_or_create_async",
+                session_manager.get_or_create,
+                key,
+            )
             extra: dict[str, Any] = {"_channel_delivery": True}
             if msg.media:
                 extra["media"] = list(msg.media)
             session.add_message("assistant", msg.content, **extra)
-            session_manager.save(session)
+            await _call_session_manager(
+                session_manager,
+                "save_async",
+                session_manager.save,
+                session,
+            )
         await bus.publish_outbound(msg)
 
     message_tool = agent.tools.get("message")
@@ -568,14 +603,14 @@ def _run_gateway(
                 if sha:
                     logger.info("Dream commit: {}", sha)
                 store.compact_history()
-                prune_dream_sessions(agent.sessions)
+                await asyncio.to_thread(prune_dream_sessions, agent.sessions)
             return None
 
         # Heartbeat is a system job that checks HEARTBEAT.md for active tasks.
         if job.name == "heartbeat":
             heartbeat_file = config.workspace_path / "HEARTBEAT.md"
             try:
-                content = heartbeat_file.read_text(encoding="utf-8")
+                content = await asyncio.to_thread(heartbeat_file.read_text, encoding="utf-8")
             except OSError:
                 logger.debug("Heartbeat: HEARTBEAT.md missing")
                 return None
@@ -583,7 +618,7 @@ def _run_gateway(
                 logger.debug("Heartbeat: HEARTBEAT.md has no active tasks")
                 return None
 
-            channel, chat_id = _pick_heartbeat_target()
+            channel, chat_id = await _pick_heartbeat_target()
             if channel == "cli":
                 return None
 
@@ -611,9 +646,19 @@ def _run_gateway(
                     message_tool.reset_suppress_delivery(suppress_token)
 
             # Keep a small tail of heartbeat history so the loop stays bounded.
-            session = agent.sessions.get_or_create("heartbeat")
+            session = await _call_session_manager(
+                agent.sessions,
+                "get_or_create_async",
+                agent.sessions.get_or_create,
+                "heartbeat",
+            )
             session.retain_recent_legal_suffix(hb_cfg.keep_recent_messages)
-            agent.sessions.save(session)
+            await _call_session_manager(
+                agent.sessions,
+                "save_async",
+                agent.sessions.save,
+                session,
+            )
 
             if not resp or not resp.content:
                 return
@@ -690,17 +735,27 @@ def _run_gateway(
         config_path=Path(config_path),
     )
 
-    def _pick_heartbeat_target() -> tuple[str, str]:
+    async def _pick_heartbeat_target() -> tuple[str, str]:
         """Pick a routable channel/chat target for heartbeat-triggered messages."""
-        sidebar_state = read_webui_sidebar_state()
+        sidebar_state = await asyncio.to_thread(read_webui_sidebar_state)
         unified_metadata = None
         if config.agents.defaults.unified_session:
-            record = session_manager.read_session_metadata(UNIFIED_SESSION_KEY)
+            record = await _call_session_manager(
+                session_manager,
+                "read_session_metadata_async",
+                session_manager.read_session_metadata,
+                UNIFIED_SESSION_KEY,
+            )
             if isinstance(record, dict) and isinstance(record.get("metadata"), dict):
                 unified_metadata = record["metadata"]
+        sessions = await _call_session_manager(
+            session_manager,
+            "list_sessions_async",
+            session_manager.list_sessions,
+        )
         return _pick_heartbeat_target_from_sessions(
             enabled_channels=channels.enabled_channels,
-            sessions=session_manager.list_sessions(),
+            sessions=sessions,
             archived_keys=sidebar_state.get("archived_keys", []),
             unified_session_metadata=unified_metadata,
         )
@@ -907,6 +962,10 @@ def _run_gateway(
                     _monitor_local_clients(),
                     name="nanobot-gateway-client-monitor",
                 ),
+                asyncio.create_task(
+                    _monitor_event_loop_lag(),
+                    name="nanobot-event-loop-lag-monitor",
+                ),
             ]
             if health_server_enabled:
                 tasks.append(asyncio.create_task(
@@ -974,7 +1033,11 @@ def _run_gateway(
                 # Flush all cached sessions to durable storage before exit.
                 # This prevents data loss on filesystems with write-back
                 # caching (rclone VFS, NFS, FUSE mounts, etc.).
-                flushed = agent.sessions.flush_all()
+                flushed = await _call_session_manager(
+                    agent.sessions,
+                    "flush_all_async",
+                    agent.sessions.flush_all,
+                )
                 if flushed:
                     logger.info("Shutdown: flushed {} session(s) to disk", flushed)
             finally:

nanobot/command/builtin.py

@@ -3,6 +3,7 @@
 from __future__ import annotations
 
 import asyncio
+import inspect
 import os
 import subprocess
 import sys
@@ -14,6 +15,8 @@ from typing import TYPE_CHECKING, Any, Literal, cast
 from nanobot import __version__
 from nanobot.bus.events import INBOUND_META_USER_SHELL, OutboundMessage
 from nanobot.command.router import CommandContext, CommandRouter, normalize_command_text
+from nanobot.session.async_compat import call_session_manager
+from nanobot.utils.cancellation import shield_and_drain
 from nanobot.utils.helpers import build_status_content
 from nanobot.utils.restart import set_restart_notice_to_env
 from nanobot.utils.workspace_prompts import initialize_workspace_prompt
@@ -22,6 +25,7 @@ if TYPE_CHECKING:
     from nanobot.agent.loop import AgentLoop
     from nanobot.session.manager import Session
     from nanobot.utils.gitstore import CommitInfo
+    from nanobot.utils.llm_runtime import LLMRuntime
 
 # WebUI protocol contract for how a slash command participates in turn state:
 # - side_channel: returns control text without starting or ending an agent turn.
@@ -201,6 +205,52 @@ def builtin_command_starts_agent_turn(text: str) -> bool:
     return spec.lifecycle == "agent_turn_with_args" and bool(args.strip())
 
 
+def _has_native_coroutine_method(target: object, name: str) -> bool:
+    """Check the real target class without trusting dynamic mock attributes."""
+    method = inspect.getattr_static(type(target), name, None)
+    return inspect.iscoroutinefunction(method)
+
+
+async def _get_or_create_session(loop: AgentLoop, key: str) -> Session:
+    sessions = loop.sessions
+    return await call_session_manager(
+        sessions,
+        "get_or_create_async",
+        sessions.get_or_create,
+        key,
+    )
+
+
+async def _save_session(loop: AgentLoop, session: Session) -> None:
+    sessions = loop.sessions
+    await call_session_manager(
+        sessions,
+        "save_async",
+        sessions.save,
+        session,
+    )
+
+
+async def _runtime_for_session(loop: AgentLoop, session: Session) -> LLMRuntime:
+    if _has_native_coroutine_method(loop, "runtime_for_session_async"):
+        return await loop.runtime_for_session_async(session)
+    return await shield_and_drain(
+        asyncio.to_thread(loop.runtime_for_session, session)
+    )
+
+
+async def _set_session_model_preset(
+    loop: AgentLoop,
+    session_key: str,
+    name: str,
+) -> LLMRuntime:
+    if _has_native_coroutine_method(loop, "set_session_model_preset_async"):
+        return await loop.set_session_model_preset_async(session_key, name)
+    return await shield_and_drain(
+        asyncio.to_thread(loop.set_session_model_preset, session_key, name)
+    )
+
+
 async def cmd_stop(ctx: CommandContext) -> OutboundMessage:
     """Cancel all active tasks and subagents for the session."""
     loop = ctx.loop
@@ -257,8 +307,8 @@ async def cmd_restart(ctx: CommandContext) -> OutboundMessage:
 async def cmd_status(ctx: CommandContext) -> OutboundMessage:
     """Build an outbound status message for a session."""
     loop = ctx.loop
-    session = ctx.session or loop.sessions.get_or_create(ctx.key)
+    session = ctx.session or await _get_or_create_session(loop, ctx.key)
-    runtime = ctx.runtime or loop.runtime_for_session(session)
+    runtime = ctx.runtime or await _runtime_for_session(loop, session)
     ctx_est = 0
     with suppress(Exception):
         ctx_est, _ = loop.consolidator.estimate_session_prompt_tokens(
@@ -306,20 +356,21 @@ async def cmd_new(ctx: CommandContext) -> OutboundMessage:
     loop = ctx.loop
     await loop._cancel_active_tasks(ctx.key)  # pyright: ignore[reportPrivateUsage]
     loop.discard_session_file_state(ctx.key)
-    session = ctx.session or loop.sessions.get_or_create(ctx.key)
+    session = ctx.session or await _get_or_create_session(loop, ctx.key)
     snapshot = list(session.messages)
     archive_snapshot = None
     runtime = None
     if session.last_consolidated < len(snapshot):
-        runtime = ctx.runtime or loop.runtime_for_session(session)
+        runtime = ctx.runtime or await _runtime_for_session(loop, session)
         archive_snapshot = replace(
             session,
             messages=snapshot,
             metadata=dict(session.metadata),
             provider_state=None,
         )
+    async def reset_and_schedule_archive() -> None:
         session.clear()
-    loop.sessions.save(session)
+        await _save_session(loop, session)
         loop.sessions.invalidate(session.key)
         if archive_snapshot is not None and runtime is not None:
             loop.schedule_background(
@@ -329,6 +380,8 @@ async def cmd_new(ctx: CommandContext) -> OutboundMessage:
                     runtime=runtime,
                 )
             )
+
+    await shield_and_drain(reset_and_schedule_archive())
     return OutboundMessage(
         channel=ctx.msg.channel, chat_id=ctx.msg.chat_id,
         content="New session started.",
@@ -377,7 +430,7 @@ async def cmd_model(ctx: CommandContext) -> OutboundMessage:
     metadata = {**dict(ctx.msg.metadata or {}), "render_as": "text"}
 
     if not args:
-        session = ctx.session or loop.sessions.get_or_create(ctx.key)
+        session = ctx.session or await _get_or_create_session(loop, ctx.key)
         return OutboundMessage(
             channel=ctx.msg.channel,
             chat_id=ctx.msg.chat_id,
@@ -387,7 +440,7 @@ async def cmd_model(ctx: CommandContext) -> OutboundMessage:
 
     name = args
     try:
-        runtime = loop.set_session_model_preset(ctx.key, name)
+        runtime = await _set_session_model_preset(loop, ctx.key, name)
     except (KeyError, ValueError) as exc:
         names = _model_preset_names(loop)
         return OutboundMessage(
@@ -848,7 +901,7 @@ async def cmd_history(ctx: CommandContext) -> OutboundMessage:
                 metadata=dict(ctx.msg.metadata or {}),
             )
 
-    session = ctx.session or ctx.loop.sessions.get_or_create(ctx.key)
+    session = ctx.session or await _get_or_create_session(ctx.loop, ctx.key)
     history = session.get_history(max_messages=0, include_runtime_context=False)
     visible = [_format_history_message(m) for m in history]
     visible = [m for m in visible if m is not None]

nanobot/cron/bound_runner.py

@@ -8,6 +8,7 @@ import time
 import uuid
 from typing import TYPE_CHECKING, Any, Protocol
 
+from nanobot.agent.automation_turns import AutomationTurnAcceptedCancellation
 from nanobot.agent.tools.cron import CronTool
 from nanobot.bus.events import InboundMessage, OutboundMessage
 from nanobot.cron.session_delivery import origin_delivery_context
@@ -127,6 +128,15 @@ async def run_bound_cron_job(
                 session_key_override=session_key,
             )
         )
+    except AutomationTurnAcceptedCancellation:
+        cron.write_run_record(
+            run_id,
+            {
+                **run_record_base,
+                "status": "accepted",
+            },
+        )
+        raise
     except (Exception, asyncio.CancelledError) as exc:
         error_text = str(exc) or exc.__class__.__name__
         cron.write_run_record(

nanobot/cron/service.py

@@ -11,11 +11,12 @@ from dataclasses import asdict
 from datetime import datetime
 from pathlib import Path
 from types import EllipsisType
-from typing import Any, Callable, Coroutine, Literal
+from typing import Any, Callable, Coroutine, Literal, TypeVar
 
 from filelock import FileLock
 from loguru import logger
 
+from nanobot.agent.automation_turns import AutomationTurnAcceptedCancellation
 from nanobot.cron.session_turns import is_bound_cron_job
 from nanobot.cron.types import (
     CronJob,
@@ -25,10 +26,14 @@ from nanobot.cron.types import (
     CronSchedule,
     CronStore,
 )
+from nanobot.utils.cancellation import shield_and_drain
 from nanobot.utils.run_records import (
     write_run_record as write_automation_run_record,
 )
 
+_FILE_LOCK_TIMEOUT_SECONDS = 5
+_T = TypeVar("_T")
+
 
 class CronJobSkippedError(Exception):
     """Raised by cron callbacks when a job was intentionally skipped."""
@@ -164,10 +169,16 @@ class CronService:
         self.store_path = store_path
         self._action_path = store_path.parent / "action.jsonl"
         self._run_records_dir = store_path.parent / "runs"
-        self._lock = FileLock(str(self._action_path.parent) + ".lock")
+        self._lock = FileLock(
+            str(self._action_path.parent) + ".lock",
+            timeout=_FILE_LOCK_TIMEOUT_SECONDS,
+        )
         self.on_job = on_job
         self._store: CronStore | None = None
         self._timer_task: asyncio.Task[None] | None = None
+        self._operation_lock = asyncio.Lock()
+        self._claimed_job_ids: set[str] = set()
+        self._event_loop: asyncio.AbstractEventLoop | None = None
         self._running = False
         self._active_executions = 0
         self._store_dirty = False
@@ -451,25 +462,58 @@ class CronService:
         """Write an internal audit record for one cron execution."""
         write_automation_run_record(self._run_records_dir, run_id, record)
 
+    async def run_sync(
+        self,
+        operation: Callable[..., _T],
+        /,
+        *args: Any,
+        **kwargs: Any,
+    ) -> _T:
+        """Serialize a complete cron transaction in a worker thread.
+
+        A running thread cannot be cancelled safely.  Keep the transaction lock
+        until it exits so cancellation is never reported while that worker can
+        still mutate cron state behind a later operation.
+        """
+        async with self._operation_lock:
+            return await shield_and_drain(
+                asyncio.to_thread(operation, *args, **kwargs)
+            )
+
     async def start(self) -> None:
-        """Start the cron service."""
+        """Start the cron service and settle accepted work before cancellation."""
+
+        async def settle_start() -> None:
+            self._event_loop = asyncio.get_running_loop()
             self._running = True
-        loaded = self._load_store()
+            try:
+                async with self._operation_lock:
+                    loaded = await asyncio.to_thread(self._load_store)
                     if loaded is None:
                         # Store file existed but was corrupt and has been preserved with
                         # a ``.corrupt-<ts>`` suffix.  Bail out instead of starting with
                         # an empty store; that would call ``_save_store`` and overwrite
                         # the now-renamed (but still recoverable) data with [].
-            self._running = False
                         raise RuntimeError(
                             f"cron store at {self.store_path} is corrupt and was preserved; "
                             "refusing to start with an empty job list. "
                             "Inspect the .corrupt-<ts> backup and restore manually."
                         )
                     self._recompute_next_runs()
-        self._save_store()
+                    await asyncio.to_thread(self._save_store)
                 self._arm_timer()
-        logger.info("Cron service started with {} jobs", len(self._store.jobs if self._store else []))
+                logger.info(
+                    "Cron service started with {} jobs",
+                    len(self._store.jobs if self._store else []),
+                )
+            except BaseException:
+                # A failed start must not retain ownership without a timer.  Caller
+                # cancellation is shielded until this composite either reaches the
+                # fully started state above or rolls back here.
+                self.stop()
+                raise
+
+        await shield_and_drain(settle_start())
 
     def stop(self) -> None:
         """Stop the cron service."""
@@ -497,8 +541,22 @@ class CronService:
                  if j.enabled and j.state.next_run_at_ms]
         return min(times) if times else None
 
+    def _request_timer_rearm(self) -> None:
+        """Re-arm on the owning event loop, including from persistence workers."""
+        if not self._running:
+            return
+        loop = self._event_loop
+        try:
+            current_loop = asyncio.get_running_loop()
+        except RuntimeError:
+            current_loop = None
+        if current_loop is loop:
+            self._arm_timer()
+        elif loop is not None and loop.is_running():
+            loop.call_soon_threadsafe(self._arm_timer)
+
     def _arm_timer(self) -> None:
-        """Schedule the next timer tick."""
+        """Schedule the next timer tick on the owning event loop."""
         if self._timer_task:
             self._timer_task.cancel()
 
@@ -520,7 +578,7 @@ class CronService:
         self._timer_task = asyncio.create_task(tick())
 
     async def _on_timer(self) -> None:
-        """Handle timer tick - run due jobs."""
+        """Run due jobs while keeping persistence transactions serialized."""
         reload_store = self._active_executions == 0
         self._active_executions += 1
         try:
@@ -528,11 +586,17 @@ class CronService:
             # to persist their advanced schedule.  Persist that exact snapshot
             # before reloading or executing anything else; otherwise the older
             # disk state can replay the same job.
+            async with self._operation_lock:
                 if self._store_dirty:
-                self._save_store()
+                    await shield_and_drain(asyncio.to_thread(self._save_store))
                     return
 
-            store = self._load_store(reload_during_execution=reload_store)
+                store = await shield_and_drain(
+                    asyncio.to_thread(
+                        self._load_store,
+                        reload_during_execution=reload_store,
+                    )
+                )
             # If a hot reload found a corrupt store on disk, ``self._store``
             # may still hold the previous, known-good in-memory snapshot.
             if store is None:
@@ -547,7 +611,8 @@ class CronService:
             for job in due_jobs:
                 await self._execute_job(job)
 
-            self._save_store()
+            async with self._operation_lock:
+                await shield_and_drain(asyncio.to_thread(self._save_store))
         except Exception:
             # A load/persist failure must not kill the scheduler: keep the
             # in-memory store and retry on the next tick.  This mirrors the
@@ -564,59 +629,125 @@ class CronService:
             # single bad tick cannot silently stop all future jobs.
             self._arm_timer()
 
-    async def _execute_job(self, job: CronJob) -> None:
+    async def _claim_job(self, job_id: str) -> bool:
-        """Execute a single job."""
+        """Claim one job without serializing callbacks for different jobs."""
-        start_ms = _now_ms()
+        async with self._operation_lock:
-        logger.info("Cron: executing job '{}' ({})", job.name, job.id)
+            if job_id in self._claimed_job_ids:
+                return False
+            self._claimed_job_ids.add(job_id)
+            return True
 
-        try:
+    async def _release_job_claim(self, job_id: str) -> None:
-            if self.on_job:
+        async with self._operation_lock:
-                await self.on_job(job)
+            self._claimed_job_ids.discard(job_id)
-
-            job.state.last_status = "ok"
-            job.state.last_error = None
-            logger.info("Cron: job '{}' completed", job.name)
-
-        except CronJobSkippedError as e:
-            job.state.last_status = "skipped"
-            job.state.last_error = str(e) or None
-            logger.warning("Cron: job '{}' skipped: {}", job.name, job.state.last_error or "")
-        except asyncio.CancelledError as e:
-            current = asyncio.current_task()
-            if current is not None and current.cancelling():
-                raise
-            job.state.last_status = "error"
-            job.state.last_error = str(e) or e.__class__.__name__
-            logger.exception("Cron: job '{}' was cancelled", job.name)
-        except Exception as e:
-            job.state.last_status = "error"
-            job.state.last_error = str(e)
-            logger.exception("Cron: job '{}' failed", job.name)
 
+    async def _settle_job_execution(
+        self,
+        job: CronJob,
+        *,
+        start_ms: int,
+        status: Literal["ok", "error", "skipped"],
+        error: str | None,
+        persist: bool = False,
+    ) -> None:
         end_ms = _now_ms()
+        async with self._operation_lock:
+            job.state.last_status = status
+            job.state.last_error = error
             job.state.last_run_at_ms = start_ms
             job.updated_at_ms = end_ms
-
             job.state.run_history.append(CronRunRecord(
                 run_at_ms=start_ms,
-            status=job.state.last_status,
+                status=status,
                 duration_ms=end_ms - start_ms,
-            error=job.state.last_error,
+                error=error,
             ))
             job.state.run_history = job.state.run_history[-self._MAX_RUN_HISTORY:]
 
-        # Handle one-shot jobs
             if job.schedule.kind == "at":
                 if job.delete_after_run:
-                store = self._require_store()
+                    store = await shield_and_drain(
+                        asyncio.to_thread(self._require_store)
+                    )
                     store.jobs = [item for item in store.jobs if item.id != job.id]
                 else:
                     job.enabled = False
                     job.state.next_run_at_ms = None
             else:
-            # Compute next run
                 job.state.next_run_at_ms = _compute_next_run(job.schedule, _now_ms())
 
+            if persist:
+                await shield_and_drain(asyncio.to_thread(self._save_store))
+
+    @staticmethod
+    async def _drain_settlement_on_cancellation(settlement: asyncio.Task[None]) -> None:
+        """Finish a short durable settlement despite repeated cancellation."""
+        while not settlement.done():
+            try:
+                await asyncio.shield(settlement)
+            except asyncio.CancelledError:
+                continue
+        settlement.result()
+
+    async def _execute_job(self, job: CronJob) -> bool:
+        """Execute a claimed job and serialize its in-memory settlement."""
+        if not await self._claim_job(job.id):
+            logger.info("Cron: job '{}' ({}) is already running", job.name, job.id)
+            return False
+
+        start_ms = _now_ms()
+        logger.info("Cron: executing job '{}' ({})", job.name, job.id)
+        status: Literal["ok", "error", "skipped"]
+        error: str | None
+        accepted_cancellation: AutomationTurnAcceptedCancellation | None = None
+
+        try:
+            try:
+                if self.on_job:
+                    await self.on_job(job)
+                status = "ok"
+                error = None
+                logger.info("Cron: job '{}' completed", job.name)
+            except AutomationTurnAcceptedCancellation as exc:
+                # The agent owns this turn now.  Advance and persist the schedule
+                # before allowing shutdown cancellation to unwind the timer.
+                status = "ok"
+                error = None
+                accepted_cancellation = exc
+                logger.info("Cron: job '{}' was accepted before cancellation", job.name)
+            except CronJobSkippedError as exc:
+                status = "skipped"
+                error = str(exc) or None
+                logger.warning("Cron: job '{}' skipped: {}", job.name, error or "")
+            except asyncio.CancelledError as exc:
+                current = asyncio.current_task()
+                if current is not None and current.cancelling():
+                    raise
+                status = "error"
+                error = str(exc) or exc.__class__.__name__
+                logger.exception("Cron: job '{}' was cancelled", job.name)
+            except Exception as exc:
+                status = "error"
+                error = str(exc)
+                logger.exception("Cron: job '{}' failed", job.name)
+
+            settlement = asyncio.create_task(
+                self._settle_job_execution(
+                    job,
+                    start_ms=start_ms,
+                    status=status,
+                    error=error,
+                    persist=accepted_cancellation is not None,
+                )
+            )
+            if accepted_cancellation is not None:
+                await self._drain_settlement_on_cancellation(settlement)
+                raise accepted_cancellation
+            await settlement
+            return True
+        finally:
+            await self._release_job_claim(job.id)
+
     def _append_action(
         self,
         action: Literal["add", "del", "update"],
@@ -697,7 +828,7 @@ class CronService:
             store = self._require_store()
             store.jobs.append(job)
             self._save_store()
-            self._arm_timer()
+            self._request_timer_rearm()
         else:
             self._append_action("add", asdict(job))
 
@@ -714,7 +845,7 @@ class CronService:
         store.jobs = [j for j in store.jobs if j.id != job.id]
         store.jobs.append(job)
         self._save_store()
-        self._arm_timer()
+        self._request_timer_rearm()
         logger.info("Cron: registered system job '{}' ({})", job.name, job.id)
         return job
 
@@ -726,7 +857,7 @@ class CronService:
         removed = len(store.jobs) < before
         if removed:
             self._save_store()
-            self._arm_timer()
+            self._request_timer_rearm()
             logger.info("Cron: removed system job {}", job_id)
         return removed
 
@@ -747,7 +878,7 @@ class CronService:
         if removed:
             if self._should_persist_store():
                 self._save_store()
-                self._arm_timer()
+                self._request_timer_rearm()
             else:
                 self._append_action("del", {"job_id": job_id})
             logger.info("Cron: removed job {}", job_id)
@@ -769,7 +900,7 @@ class CronService:
                     job.state.next_run_at_ms = None
                 if self._should_persist_store():
                     self._save_store()
-                    self._arm_timer()
+                    self._request_timer_rearm()
                 else:
                     self._append_action("update", asdict(job))
                 return job
@@ -825,7 +956,7 @@ class CronService:
 
         if self._should_persist_store():
             self._save_store()
-            self._arm_timer()
+            self._request_timer_rearm()
         else:
             self._append_action("update", asdict(job))
 
@@ -840,19 +971,31 @@ class CronService:
             # A manual run is another side-effecting entrypoint.  Do not start
             # it while the result of a previous timer execution is still only
             # in memory.
+            async with self._operation_lock:
                 if self._store_dirty:
-                self._save_store()
+                    await shield_and_drain(asyncio.to_thread(self._save_store))
-            store = self._require_store(reload_during_execution=reload_store)
+                store = await shield_and_drain(
+                    asyncio.to_thread(
+                        self._require_store,
+                        reload_during_execution=reload_store,
+                    )
+                )
             for job in store.jobs:
                 if job.id == job_id:
                     if self._is_unbound_agent_job(job):
+                        async with self._operation_lock:
                             self._enforce_agent_binding(job)
-                        self._save_store()
+                            await shield_and_drain(
+                                asyncio.to_thread(self._save_store)
+                            )
                         return False
                     if not force and not job.enabled:
                         return False
-                    await self._execute_job(job)
+                    executed = await self._execute_job(job)
-                    self._save_store()
+                    if not executed:
+                        return False
+                    async with self._operation_lock:
+                        await shield_and_drain(asyncio.to_thread(self._save_store))
                     return True
             return False
         finally:

nanobot/providers/image_generation.py

@@ -20,7 +20,7 @@ from nanobot.providers.registry import find_by_name
 from nanobot.security.network import (
     PinnedDNSAsyncTransport,
     UnsafeURLRequestError,
-    resolve_url_target,
+    async_resolve_url_target,
 )
 from nanobot.utils.helpers import detect_image_mime
 
@@ -174,7 +174,7 @@ async def _download_image_data_url(
             current_url = url
             for _ in range(_IMAGE_DOWNLOAD_MAX_REDIRECTS + 1):
                 if proxy:
-                    ok, error, _ = resolve_url_target(
+                    ok, error, _ = await async_resolve_url_target(
                         current_url,
                         trust_remote_dns=True,
                     )

nanobot/sdk/clients.py

@@ -210,18 +210,20 @@ class RuntimeClient:
 
     async def compact_session(self, session_key: str) -> SessionSnapshot:
         """Run token consolidation for one session."""
-        session = self._loop.sessions.get_or_create(session_key)
+        session = await self._loop.sessions.get_or_create_async(session_key)
-        runtime = self._loop.runtime_for_session(session)
+        runtime = await self._loop.runtime_for_session_async(session)
         await self._loop.consolidator.maybe_consolidate_by_tokens(
             session,
             runtime=runtime,
         )
-        return snapshot_from_session(self._loop.sessions.get_or_create(session_key))
+        return snapshot_from_session(
+            await self._loop.sessions.get_or_create_async(session_key)
+        )
 
     async def compact_idle_session(self, session_key: str, *, max_suffix: int = 8) -> str | None:
         """Run idle-session compaction for one session and return the summary."""
-        session = self._loop.sessions.get_or_create(session_key)
+        session = await self._loop.sessions.get_or_create_async(session_key)
-        runtime = self._loop.runtime_for_session(session)
+        runtime = await self._loop.runtime_for_session_async(session)
         return await self._loop.consolidator.compact_idle_session(
             session_key,
             runtime=runtime,

nanobot/security/network.py

@@ -29,6 +29,7 @@ _BLOCKED_NETWORKS = [
 
 _URL_RE = re.compile(r"https?://[^\s\"'`;|<>]+", re.IGNORECASE)
 _allowed_networks: list[ipaddress.IPv4Network | ipaddress.IPv6Network] = []
+_DNS_RESOLUTION_TIMEOUT_SECONDS = 5.0
 
 
 def is_loopback_host(host: str) -> bool:
@@ -75,6 +76,63 @@ def _is_private(addr: ipaddress.IPv4Address | ipaddress.IPv6Address) -> bool:
     return any(normalized in net for net in _BLOCKED_NETWORKS)
 
 
+def _parse_url_hostname(url: str) -> tuple[str | None, str | None]:
+    try:
+        parsed = urlparse(url)
+    except Exception as exc:
+        return None, str(exc)
+    if parsed.scheme not in ("http", "https"):
+        return None, f"Only http/https allowed, got '{parsed.scheme or 'none'}'"
+    if not parsed.netloc:
+        return None, "Missing domain"
+    if not parsed.hostname:
+        return None, "Missing hostname"
+    return parsed.hostname, None
+
+
+def _unresolved_target_result(
+    hostname: str,
+    *,
+    trust_remote_dns: bool,
+) -> tuple[bool, str, tuple[str, ...]]:
+    if not trust_remote_dns:
+        return False, f"Cannot resolve hostname: {hostname}", ()
+
+    normalized_hostname = hostname.rstrip(".").lower()
+    if normalized_hostname == "localhost" or normalized_hostname.endswith(".localhost"):
+        return False, f"Blocked local/internal hostname: {hostname}", ()
+
+    try:
+        literal_addr = ipaddress.ip_address(normalized_hostname)
+    except ValueError:
+        return True, "", ()
+    if _is_private(literal_addr):
+        return False, f"Blocked private/internal address: {literal_addr}", ()
+    return True, "", (str(_normalize_addr(literal_addr)),)
+
+
+def _resolved_target_result(
+    hostname: str,
+    infos: list[Any],
+    *,
+    allow_loopback: bool,
+) -> tuple[bool, str, tuple[str, ...]]:
+    addrs: list[ipaddress.IPv4Address | ipaddress.IPv6Address] = []
+    for info in infos:
+        try:
+            addr = ipaddress.ip_address(info[4][0])
+        except (IndexError, TypeError, ValueError):
+            continue
+        addrs.append(addr)
+    if allow_loopback and _is_allowed_loopback_target(hostname, addrs):
+        return True, "", tuple(dict.fromkeys(str(_normalize_addr(addr)) for addr in addrs))
+    for addr in addrs:
+        if _is_private(addr):
+            return False, f"Blocked: {hostname} resolves to private/internal address {addr}", ()
+
+    return True, "", tuple(dict.fromkeys(str(_normalize_addr(addr)) for addr in addrs))
+
+
 def resolve_url_target(
     url: str,
     *,
@@ -97,52 +155,43 @@ def resolve_url_target(
     resolved_ips contains the public IPs that were validated for this URL, or
     is empty when an unresolved hostname is delegated to a trusted proxy.
     """
-    try:
+    hostname, error = _parse_url_hostname(url)
-        p = urlparse(url)
+    if hostname is None:
-    except Exception as e:
+        return False, error or "Missing hostname", ()
-        return False, str(e), ()
-
-    if p.scheme not in ("http", "https"):
-        return False, f"Only http/https allowed, got '{p.scheme or 'none'}'", ()
-    if not p.netloc:
-        return False, "Missing domain", ()
-
-    hostname = p.hostname
-    if not hostname:
-        return False, "Missing hostname", ()
-
     try:
         infos = socket.getaddrinfo(hostname, None, socket.AF_UNSPEC, socket.SOCK_STREAM)
     except socket.gaierror:
-        if not trust_remote_dns:
+        return _unresolved_target_result(hostname, trust_remote_dns=trust_remote_dns)
-            return False, f"Cannot resolve hostname: {hostname}", ()
+    return _resolved_target_result(hostname, infos, allow_loopback=allow_loopback)
 
-        normalized_hostname = hostname.rstrip(".").lower()
-        if normalized_hostname == "localhost" or normalized_hostname.endswith(".localhost"):
-            return False, f"Blocked local/internal hostname: {hostname}", ()
 
+async def async_resolve_url_target(
+    url: str,
+    *,
+    allow_loopback: bool = False,
+    trust_remote_dns: bool = False,
+    timeout_s: float = _DNS_RESOLUTION_TIMEOUT_SECONDS,
+) -> tuple[bool, str, tuple[str, ...]]:
+    """Resolve and validate an HTTP target without blocking the event loop."""
+    hostname, error = _parse_url_hostname(url)
+    if hostname is None:
+        return False, error or "Missing hostname", ()
+    loop = asyncio.get_running_loop()
     try:
-            literal_addr = ipaddress.ip_address(normalized_hostname)
+        infos = await asyncio.wait_for(
-        except ValueError:
+            loop.getaddrinfo(
-            return True, "", ()
+                hostname,
-        if _is_private(literal_addr):
+                None,
-            return False, f"Blocked private/internal address: {literal_addr}", ()
+                family=socket.AF_UNSPEC,
-        return True, "", (str(_normalize_addr(literal_addr)),)
+                type=socket.SOCK_STREAM,
-
+            ),
-    addrs: list[ipaddress.IPv4Address | ipaddress.IPv6Address] = []
+            timeout=timeout_s,
-    for info in infos:
+        )
-        try:
+    except asyncio.TimeoutError:
-            addr = ipaddress.ip_address(info[4][0])
+        return False, f"Timed out resolving hostname: {hostname}", ()
-        except ValueError:
+    except socket.gaierror:
-            continue
+        return _unresolved_target_result(hostname, trust_remote_dns=trust_remote_dns)
-        addrs.append(addr)
+    return _resolved_target_result(hostname, infos, allow_loopback=allow_loopback)
-    if allow_loopback and _is_allowed_loopback_target(hostname, addrs):
-        return True, "", tuple(dict.fromkeys(str(_normalize_addr(addr)) for addr in addrs))
-    for addr in addrs:
-        if _is_private(addr):
-            return False, f"Blocked: {hostname} resolves to private/internal address {addr}", ()
-
-    return True, "", tuple(dict.fromkeys(str(_normalize_addr(addr)) for addr in addrs))
 
 
 def validate_url_target(url: str, *, allow_loopback: bool = False) -> tuple[bool, str]:
@@ -151,6 +200,16 @@ def validate_url_target(url: str, *, allow_loopback: bool = False) -> tuple[bool
     return ok, error
 
 
+async def async_validate_url_target(
+    url: str,
+    *,
+    allow_loopback: bool = False,
+) -> tuple[bool, str]:
+    """Validate a URL using the event loop's asynchronous resolver."""
+    ok, error, _ = await async_resolve_url_target(url, allow_loopback=allow_loopback)
+    return ok, error
+
+
 def env_proxy_applies_to_url(url: str) -> bool:
     """Return True when process proxy settings would proxy this URL."""
     try:
@@ -277,7 +336,10 @@ class PinnedDNSAsyncTransport(httpx.AsyncBaseTransport):
 
     async def handle_async_request(self, request: httpx.Request) -> httpx.Response:
         url = str(request.url)
-        ok, error, resolved_ips = resolve_url_target(url, allow_loopback=self._allow_loopback)
+        ok, error, resolved_ips = await async_resolve_url_target(
+            url,
+            allow_loopback=self._allow_loopback,
+        )
         if not ok:
             raise UnsafeURLRequestError(error, request=request)
         async with self._resolver_lock:
@@ -320,6 +382,46 @@ def validate_resolved_url(url: str) -> tuple[bool, str]:
     return True, ""
 
 
+async def async_validate_resolved_url(url: str) -> tuple[bool, str]:
+    """Validate a redirect target without blocking on domain resolution."""
+    try:
+        parsed = urlparse(url)
+    except Exception:
+        return True, ""
+    hostname = parsed.hostname
+    if not hostname:
+        return True, ""
+    try:
+        addr = ipaddress.ip_address(hostname)
+    except ValueError:
+        loop = asyncio.get_running_loop()
+        try:
+            infos = await asyncio.wait_for(
+                loop.getaddrinfo(
+                    hostname,
+                    None,
+                    family=socket.AF_UNSPEC,
+                    type=socket.SOCK_STREAM,
+                ),
+                timeout=_DNS_RESOLUTION_TIMEOUT_SECONDS,
+            )
+        except asyncio.TimeoutError:
+            return False, f"Timed out resolving redirect hostname: {hostname}"
+        except socket.gaierror:
+            return True, ""
+        for info in infos:
+            try:
+                addr = ipaddress.ip_address(info[4][0])
+            except (IndexError, TypeError, ValueError):
+                continue
+            if _is_private(addr):
+                return False, f"Redirect target {hostname} resolves to private address {addr}"
+        return True, ""
+    if _is_private(addr):
+        return False, f"Redirect target is a private address: {addr}"
+    return True, ""
+
+
 def contains_internal_url(command: str, *, allow_loopback: bool = False) -> bool:
     """Return True if the command string contains a URL targeting an internal/private address."""
     for m in _URL_RE.finditer(command):

nanobot/session/async_compat.py

@@ -0,0 +1,29 @@
+"""Compatibility bridge for asynchronous SessionManager operations."""
+
+import asyncio
+import inspect
+from collections.abc import Awaitable, Callable
+from typing import Any, TypeVar, cast
+
+from nanobot.utils.cancellation import shield_and_drain
+
+_SessionResult = TypeVar("_SessionResult")
+
+
+async def call_session_manager(
+    manager: object,
+    async_method_name: str,
+    sync_method: Callable[..., _SessionResult],
+    /,
+    *args: Any,
+    **kwargs: Any,
+) -> _SessionResult:
+    """Prefer a class-declared coroutine, or offload the established sync contract."""
+    class_async_method = inspect.getattr_static(type(manager), async_method_name, None)
+    if inspect.iscoroutinefunction(class_async_method):
+        async_method = cast(
+            Callable[..., Awaitable[_SessionResult]],
+            getattr(manager, async_method_name),
+        )
+        return await async_method(*args, **kwargs)
+    return await shield_and_drain(asyncio.to_thread(sync_method, *args, **kwargs))

nanobot/session/manager.py

@@ -1,5 +1,6 @@
 """Session management for conversation history."""
 
+import asyncio
 import base64
 import errno
 import hashlib
@@ -28,6 +29,7 @@ from nanobot.runtime_context import (
     public_history_message,
 )
 from nanobot.session.model_selection import SESSION_MODEL_PRESET_METADATA_KEY
+from nanobot.utils.cancellation import shield_and_drain
 from nanobot.utils.helpers import (
     content_with_media_breadcrumbs,
     ensure_dir,
@@ -71,6 +73,7 @@ _WORKSPACE_STATE_DIR = ".nanobot"
 _WORKSPACE_ID_FILE = "workspace-id"
 _WORKSPACE_ID_RE = re.compile(r"^[0-9a-f]{32}$")
 _SESSION_MIGRATION_LOCK_TIMEOUT_SECONDS = 30
+_SESSION_FILES_LOCK_TIMEOUT_SECONDS = 5
 _SESSION_FILES_LOCK_FILENAME = ".session-files.lock"
 _COPY_CHUNK_SIZE = 1024 * 1024
 
@@ -560,7 +563,8 @@ class JsonlSessionStore:
             self.sessions_dir = ensure_dir(root / workspace_id)
             self.legacy_sessions_dir = get_legacy_sessions_dir()
             self._session_files_lock = FileLock(
-                str(self.sessions_dir / _SESSION_FILES_LOCK_FILENAME)
+                str(self.sessions_dir / _SESSION_FILES_LOCK_FILENAME),
+                timeout=_SESSION_FILES_LOCK_TIMEOUT_SECONDS,
             )
             with self._session_files_lock:
                 self._migrate_from_workspace(canonical_workspace)
@@ -1642,6 +1646,7 @@ class SessionManager:
         self._cache: OrderedDict[str, Session] = OrderedDict()
         # Preserve identity for sessions held by active callers without retaining idle ones.
         self._overflow_cache: WeakValueDictionary[str, Session] = WeakValueDictionary()
+        self._async_session_locks: WeakValueDictionary[str, asyncio.Lock] = WeakValueDictionary()
         self._max_cached_sessions = SESSION_CACHE_MAX_SIZE
         self._delete_observer: Callable[[str], None] | None = None
 
@@ -1741,6 +1746,28 @@ class SessionManager:
         self._remember(session)
         return session
 
+    def _async_session_lock(self, key: str) -> asyncio.Lock:
+        lock = self._async_session_locks.get(key)
+        if lock is None:
+            lock = asyncio.Lock()
+            self._async_session_locks[key] = lock
+        return lock
+
+    async def get_or_create_async(self, key: str) -> Session:
+        """Load a session without running file I/O or lock waits on the event loop."""
+        cached = self.get_cached(key)
+        if cached is not None:
+            return cached
+        async with self._async_session_lock(key):
+            cached = self.get_cached(key)
+            if cached is not None:
+                return cached
+            session = await asyncio.to_thread(self._load, key)
+            if session is None:
+                session = Session(key=key)
+            self._remember(session)
+            return session
+
     def get_or_create_transient(
         self,
         key: str,
@@ -1774,6 +1801,17 @@ class SessionManager:
         self._store.save(session, fsync=fsync)
         self._remember(session)
 
+    async def save_async(self, session: Session, *, fsync: bool = False) -> None:
+        """Persist a session without blocking the caller's event loop."""
+        if not session.policy.persist:
+            return
+
+        async def save_and_remember() -> None:
+            await asyncio.to_thread(self._store.save, session, fsync=fsync)
+            self._remember(session)
+
+        await shield_and_drain(save_and_remember())
+
     def save_runtime_checkpoint(self, session: Session) -> None:
         """Persist volatile recovery state without rewriting long history."""
         if not session.policy.persist:
@@ -1786,6 +1824,23 @@ class SessionManager:
         # they opt into a dedicated checkpoint primitive.
         self.save(session)
 
+    async def save_runtime_checkpoint_async(self, session: Session) -> None:
+        """Persist an in-flight checkpoint without blocking the event loop."""
+        if not session.policy.persist:
+            return
+
+        async def save_and_remember() -> None:
+            if self._store is self._jsonl_store:
+                await asyncio.to_thread(
+                    self._jsonl_store.save_runtime_checkpoint,
+                    session,
+                )
+            else:
+                await asyncio.to_thread(self._store.save, session)
+            self._remember(session)
+
+        await shield_and_drain(save_and_remember())
+
     def rename_model_preset(self, old_name: str, new_name: str) -> int:
         """Rename a session-scoped model preset across durable and live sessions."""
         if old_name == new_name:
@@ -1827,6 +1882,21 @@ class SessionManager:
             raise
         return len(changed)
 
+    async def flush_all_async(self) -> int:
+        """Re-save every cached session without blocking the event loop."""
+        cached = dict(self._overflow_cache.items())
+        cached.update(self._cache)
+        flushed = 0
+        for key, session in cached.items():
+            try:
+                await shield_and_drain(
+                    asyncio.to_thread(self._store.save, session, fsync=True)
+                )
+                flushed += 1
+            except Exception:
+                logger.warning("Failed to flush session {}", key, exc_info=True)
+        return flushed
+
     def flush_all(self) -> int:
         """Re-save every cached session with fsync for durable shutdown.
 
@@ -1858,6 +1928,18 @@ class SessionManager:
             self._delete_observer(key)
         return deleted
 
+    async def delete_session_async(self, key: str) -> bool:
+        """Delete a session without blocking the event loop."""
+
+        async def delete_and_notify() -> bool:
+            self.invalidate(key)
+            deleted = await asyncio.to_thread(self._store.delete, key)
+            if self._delete_observer is not None:
+                self._delete_observer(key)
+            return deleted
+
+        return await shield_and_drain(delete_and_notify())
+
     def restore_sessions_to_workspace(self) -> SessionRestoreResult:
         """Restore session files to the pre-relocation path for an explicit rollback."""
         return self._jsonl_store.restore_to_workspace()
@@ -1930,6 +2012,10 @@ class SessionManager:
         """Read session metadata without loading the transcript."""
         return cast(dict[str, Any] | None, self._store.read_metadata(key))
 
+    async def read_session_metadata_async(self, key: str) -> dict[str, Any] | None:
+        """Read session metadata without blocking the event loop."""
+        return await asyncio.to_thread(self.read_session_metadata, key)
+
     def update_session_metadata(
         self,
         key: str,
@@ -1943,5 +2029,31 @@ class SessionManager:
             session.metadata.update(deepcopy(updates))
         return updated
 
+    async def update_session_metadata_async(
+        self,
+        key: str,
+        updates: dict[str, Any],
+        *,
+        fsync: bool = False,
+    ) -> bool:
+        """Update metadata without blocking the event loop."""
+
+        async def update_and_refresh_cache() -> bool:
+            updated = await asyncio.to_thread(
+                self._store.update_metadata,
+                key,
+                updates,
+                fsync=fsync,
+            )
+            if updated and (session := self.get_cached(key)) is not None:
+                session.metadata.update(deepcopy(updates))
+            return updated
+
+        return await shield_and_drain(update_and_refresh_cache())
+
     def list_sessions(self) -> list[dict[str, Any]]:
         return cast(list[dict[str, Any]], self._store.list_sessions())
+
+    async def list_sessions_async(self) -> list[dict[str, Any]]:
+        """List persisted sessions without blocking the event loop."""
+        return await asyncio.to_thread(self.list_sessions)

nanobot/session/recovery.py

@@ -25,6 +25,7 @@ from nanobot.bus.outbound_events import (
 )
 from nanobot.bus.queue import MessageBus
 from nanobot.session import turn_continuation
+from nanobot.session.async_compat import call_session_manager
 from nanobot.session.keys import UNIFIED_SESSION_KEY, last_channel_from_metadata
 from nanobot.session.manager import Session, SessionManager
 from nanobot.webui.metadata import WEBUI_TURN_METADATA_KEY
@@ -460,6 +461,37 @@ class RecoveryCoordinator:
         repr=False,
     )
 
+    async def _get_or_create_session(self, key: str) -> Session:
+        return await call_session_manager(
+            self.sessions,
+            "get_or_create_async",
+            self.sessions.get_or_create,
+            key,
+        )
+
+    async def _save_session(self, session: Session) -> None:
+        await call_session_manager(
+            self.sessions,
+            "save_async",
+            self.sessions.save,
+            session,
+        )
+
+    async def _read_session_metadata(self, key: str) -> dict[str, Any] | None:
+        return await call_session_manager(
+            self.sessions,
+            "read_session_metadata_async",
+            self.sessions.read_session_metadata,
+            key,
+        )
+
+    async def _list_sessions(self) -> list[dict[str, Any]]:
+        return await call_session_manager(
+            self.sessions,
+            "list_sessions_async",
+            self.sessions.list_sessions,
+        )
+
     def register_recovery_task(self, session_key: str, task: asyncio.Task[Any]) -> None:
         """Track the task that owns an explicit recovery continuation."""
         self._active_recovery_tasks[session_key] = task
@@ -482,8 +514,8 @@ class RecoveryCoordinator:
 
     async def scan(self) -> None:
         """Recover every interrupted WebUI session once at gateway startup."""
-        for key in self._recovery_candidates():
+        for key in await self._recovery_candidates():
-            metadata_payload = self.sessions.read_session_metadata(key)
+            metadata_payload = await self._read_session_metadata(key)
             raw_metadata = metadata_payload.get("metadata") if metadata_payload else None
             metadata = cast(dict[str, Any], raw_metadata) if isinstance(raw_metadata, dict) else {}
             route = self._websocket_route_for(key, metadata)
@@ -492,7 +524,7 @@ class RecoveryCoordinator:
             unfinished = self._has_unfinished_webui_transcript(key)
             if not self._needs_recovery(metadata) and not unfinished:
                 continue
-            session = self.sessions.get_or_create(key)
+            session = await self._get_or_create_session(key)
             try:
                 await self._recover_session(session, route[1])
                 await self._requeue_pending_followups(session)
@@ -507,14 +539,14 @@ class RecoveryCoordinator:
                     reason="recovery_failed",
                     can_continue=False,
                 )
-                self.sessions.save(session)
+                await self._save_session(session)
                 await self._publish(route[1], failed)
 
-    def _recovery_candidates(self) -> list[str]:
+    async def _recovery_candidates(self) -> list[str]:
         """Discover canonical and transcript-only WebUI sessions cheaply."""
         candidates = dict.fromkeys(
             key
-            for item in self.sessions.list_sessions()
+            for item in await self._list_sessions()
             if isinstance((key := item.get("key")), str)
         )
         try:
@@ -523,7 +555,7 @@ class RecoveryCoordinator:
             # duplicating its filename and migration rules here would drift.
             from nanobot.webui.session_list_index import list_webui_sessions
 
-            for item in list_webui_sessions(self.sessions):
+            for item in await asyncio.to_thread(list_webui_sessions, self.sessions):
                 key = item.get("key")
                 if isinstance(key, str):
                     candidates.setdefault(key, None)
@@ -549,7 +581,7 @@ class RecoveryCoordinator:
         """Reject stale queued recoveries and let new user input supersede them."""
         recovery_id = message.metadata.get(RECOVERY_INBOUND_METADATA_KEY)
         if isinstance(recovery_id, str):
-            session = self.sessions.get_or_create(message.session_key)
+            session = await self._get_or_create_session(message.session_key)
             state = recovery_state_from_metadata(session.metadata)
             return bool(
                 state
@@ -558,7 +590,7 @@ class RecoveryCoordinator:
             )
         if message.channel != "websocket":
             return True
-        session = self.sessions.get_or_create(message.session_key)
+        session = await self._get_or_create_session(message.session_key)
         state = recovery_state_from_metadata(session.metadata)
         if state and state["status"] in {"resuming", "awaiting_user", "failed"}:
             await self._cancel_active_recovery(message.session_key)
@@ -572,13 +604,13 @@ class RecoveryCoordinator:
                 attempts=cast(int, state.get("attempts", 0)),
                 reason="superseded",
             )
-            self.sessions.save(session)
+            await self._save_session(session)
             await self._publish(message.chat_id, recovered)
         return True
 
     async def turn_completed(self, session_key: str) -> None:
         """Resolve a resuming state after the recovered turn commits."""
-        session = self.sessions.get_or_create(session_key)
+        session = await self._get_or_create_session(session_key)
         state = recovery_state_from_metadata(session.metadata)
         if not state or state["status"] != "resuming":
             return
@@ -592,7 +624,7 @@ class RecoveryCoordinator:
             attempts=cast(int, state.get("attempts", 0)),
             reason="continued",
         )
-        self.sessions.save(session)
+        await self._save_session(session)
         await self._publish(route[1], recovered)
 
     async def handle_action(self, action: str, payload: dict[str, Any]) -> dict[str, Any]:
@@ -603,7 +635,7 @@ class RecoveryCoordinator:
             raise RecoveryActionError("missing chat_id")
         if not isinstance(recovery_id, str) or not recovery_id:
             raise RecoveryActionError("missing recovery_id")
-        session = self.sessions.get_or_create(self._session_key(chat_id))
+        session = await self._get_or_create_session(self._session_key(chat_id))
         state = recovery_state_from_metadata(session.metadata)
         if not state or state["recovery_id"] != recovery_id:
             raise RecoveryActionError("recovery state is stale", status=409)
@@ -618,7 +650,7 @@ class RecoveryCoordinator:
                 attempts=cast(int, state.get("attempts", 0)),
                 reason="dismissed",
             )
-            self.sessions.save(session)
+            await self._save_session(session)
             await self._publish(chat_id, next_state)
             return next_state
         if action != "continue":
@@ -635,7 +667,7 @@ class RecoveryCoordinator:
             reason="user_confirmed",
             resume_message_count=len(session.messages),
         )
-        self.sessions.save(session)
+        await self._save_session(session)
         await self._publish(chat_id, next_state)
         await self._queue_continuation(session, chat_id, next_state)
         return next_state
@@ -668,7 +700,7 @@ class RecoveryCoordinator:
                         attempts=cast(int, state.get("attempts", 1)),
                         reason="loop_guard",
                     )
-                self.sessions.save(session)
+                await self._save_session(session)
                 await self._publish(chat_id, next_state)
             elif self._has_unfinished_webui_transcript(session.key):
                 # A normal last-client shutdown can materialize the checkpoint
@@ -690,7 +722,7 @@ class RecoveryCoordinator:
                     ),
                     can_continue=can_continue,
                 )
-                self.sessions.save(session)
+                await self._save_session(session)
                 await self._publish(chat_id, waiting)
             return
         if state and state["status"] in {"awaiting_user", "failed"}:
@@ -706,7 +738,7 @@ class RecoveryCoordinator:
                 attempts=cast(int, state.get("attempts", 1)),
                 reason="loop_guard",
             )
-            self.sessions.save(session)
+            await self._save_session(session)
             await self._publish(chat_id, waiting)
             return
 
@@ -724,7 +756,7 @@ class RecoveryCoordinator:
                 reason="checkpoint_unknown",
                 can_continue=False,
             )
-            self.sessions.save(session)
+            await self._save_session(session)
             await self._publish(chat_id, waiting)
             return
         if checkpoint is not None and not _runtime_checkpoint_is_well_formed(checkpoint):
@@ -738,7 +770,7 @@ class RecoveryCoordinator:
                 reason="checkpoint_invalid",
                 can_continue=False,
             )
-            self.sessions.save(session)
+            await self._save_session(session)
             await self._publish(chat_id, waiting)
             return
         if phase == "final_response":
@@ -750,7 +782,7 @@ class RecoveryCoordinator:
                 attempts=0,
                 reason="answer_restored",
             )
-            self.sessions.save(session)
+            await self._save_session(session)
             await self._publish(chat_id, recovered)
             return
         if phase in _UNCERTAIN_TOOL_PHASES or pending_calls:
@@ -762,7 +794,7 @@ class RecoveryCoordinator:
                 attempts=0,
                 reason="tool_state_unknown",
             )
-            self.sessions.save(session)
+            await self._save_session(session)
             await self._publish(chat_id, waiting)
             return
         # A gateway restart is a lifecycle boundary.  Never enqueue model work
@@ -777,7 +809,7 @@ class RecoveryCoordinator:
             attempts=0,
             reason="restart_requires_confirmation",
         )
-        self.sessions.save(session)
+        await self._save_session(session)
         await self._publish(chat_id, waiting)
 
     async def _queue_continuation(

nanobot/session/webui_turns.py

@@ -2,6 +2,7 @@
 
 from __future__ import annotations
 
+import asyncio
 import re
 import time
 from collections.abc import Awaitable, Callable
@@ -176,7 +177,7 @@ async def maybe_generate_webui_title(
     model: str,
 ) -> bool:
     """Generate and persist a short title for WebUI-owned sessions only."""
-    session = sessions.get_or_create(session_key)
+    session = await sessions.get_or_create_async(session_key)
     if session.metadata.get(WEBUI_SESSION_METADATA_KEY) is not True:
         return False
     if session.metadata.get(WEBUI_TITLE_USER_EDITED_METADATA_KEY) is True:
@@ -187,7 +188,7 @@ async def maybe_generate_webui_title(
         if cleaned_current_title:
             if cleaned_current_title != current_title:
                 session.metadata[WEBUI_TITLE_METADATA_KEY] = cleaned_current_title
-                sessions.save(session)
+                await sessions.save_async(session)
             return False
         session.metadata.pop(WEBUI_TITLE_METADATA_KEY, None)
 
@@ -241,7 +242,7 @@ async def maybe_generate_webui_title(
         )
         return False
     session.metadata[WEBUI_TITLE_METADATA_KEY] = title
-    sessions.save(session)
+    await sessions.save_async(session)
     return True
 
 
@@ -437,8 +438,8 @@ class WebuiTurnRoutePolicy:
             )
             and route.channel == "websocket"
         ):
-            session = self.sessions.get_or_create(session_key)
+            session = self.sessions.get_cached(session_key)
-            if session.metadata.get(WEBUI_SESSION_METADATA_KEY) is True:
+            if session is not None and session.metadata.get(WEBUI_SESSION_METADATA_KEY) is True:
                 metadata = dict(route.metadata)
                 turn_prefix = "session-input" if internal_user_input else "subagent"
                 metadata.update({
@@ -580,7 +581,7 @@ class WebuiTurnCoordinator:
             or not session_key.startswith("websocket:")
         ):
             return
-        persisted = self.sessions.read_session_metadata(session_key)
+        persisted = await self.sessions.read_session_metadata_async(session_key)
         metadata_value: object = persisted.get("metadata") if persisted is not None else None
         metadata = (
             cast(dict[str, Any], metadata_value)
@@ -591,7 +592,8 @@ class WebuiTurnCoordinator:
             return
         public_metadata = _session_message_public_metadata(envelope)
         try:
-            append_session_message_input(
+            await asyncio.to_thread(
+                append_session_message_input,
                 session_key,
                 content=event.content,
                 created_at_ms=envelope["created_at_ms"],
@@ -616,7 +618,8 @@ class WebuiTurnCoordinator:
     def _handle_session_turn_started(self, event: SessionTurnStarted) -> None:
         if not self._is_websocket_event(event.context):
             return
-        session = self.sessions.get_or_create(event.context.session_key)
+        session = self.sessions.get_cached(event.context.session_key)
+        if session is not None:
             mark_webui_session(session, event.context.metadata)
 
     async def _handle_run_status_changed(self, event: TurnRunStatusChanged) -> None:
@@ -703,7 +706,7 @@ class WebuiTurnCoordinator:
         if msg.channel != "websocket":
             return
 
-        session = self.sessions.get_or_create(session_key)
+        session = await self.sessions.get_or_create_async(session_key)
         await self.bus.publish_outbound(
             outbound_message_for_event(
                 channel=msg.channel,

nanobot/triggers/local_runner.py

@@ -5,17 +5,24 @@ from __future__ import annotations
 import asyncio
 import uuid
 from collections.abc import Awaitable, Callable
-from typing import Any
+from contextlib import suppress
+from typing import Any, TypeVar
 
 from loguru import logger
 
-from nanobot.agent.automation_turns import AutomationTurnError
+from nanobot.agent.automation_turns import (
+    AutomationTurnAcceptedCancellation,
+    AutomationTurnError,
+)
 from nanobot.bus.events import InboundMessage, OutboundMessage
 from nanobot.triggers.local_session_turns import LOCAL_TRIGGER_META
 from nanobot.triggers.local_store import LocalTriggerStore
 from nanobot.triggers.local_types import LocalTrigger, TriggerDelivery
+from nanobot.utils.cancellation import shield_and_drain
 from nanobot.webui.metadata import WEBUI_MESSAGE_SOURCE_METADATA_KEY, WEBUI_TURN_METADATA_KEY
 
+_T = TypeVar("_T")
+
 
 async def run_local_trigger_queue(
     *,
@@ -29,14 +36,16 @@ async def run_local_trigger_queue(
     if submit_turn is None:
         raise ValueError("run_local_trigger_queue requires submit_turn")
     logger.info("Local trigger queue started")
-    recovered = store.recover_processing_deliveries()
+    recovered = await shield_and_drain(asyncio.to_thread(store.recover_processing_deliveries))
     if recovered:
         logger.warning(
             "Trigger: recovered {} interrupted delivery file(s) from processing",
             recovered,
         )
     while True:
-        deliveries = store.claim_deliveries(limit=batch_size)
+        deliveries = await shield_and_drain(
+            asyncio.to_thread(store.claim_deliveries, limit=batch_size)
+        )
         if not deliveries:
             await asyncio.sleep(poll_interval_s)
             continue
@@ -49,30 +58,24 @@ async def run_local_trigger_queue(
                     submit_turn=submit_turn,
                     is_channel_enabled=is_channel_enabled,
                 )
-                store.complete_delivery(delivery)
+            except _DeliverySettledOnCancellation:
+                raise
             except asyncio.CancelledError as exc:
-                store.retry_delivery(delivery, str(exc) or exc.__class__.__name__)
+                error = str(exc) or exc.__class__.__name__
-                _write_delivery_run_record(
+                await shield_and_drain(asyncio.to_thread(store.retry_delivery, delivery, error))
+                await _write_delivery_run_record(
                     store,
                     delivery,
                     status="interrupted",
-                    error=str(exc) or exc.__class__.__name__,
+                    error=error,
                 )
                 raise
             except _TerminalDeliveryError as exc:
-                store.record_delivery(
+                await _await_delivery_settlement(
-                    delivery.trigger_id,
+                    _settle_failed_delivery(store, delivery, error=str(exc)),
-                    status="error",
+                    store=store,
-                    error=str(exc),
+                    delivery=delivery,
-                    run_at_ms=delivery.created_at_ms,
                 )
-                _write_delivery_run_record(
-                    store,
-                    delivery,
-                    status="error",
-                    error=str(exc),
-                )
-                store.complete_delivery(delivery)
                 logger.warning(
                     "Trigger: dropped delivery {} for {}: {}",
                     delivery.id,
@@ -81,19 +84,11 @@ async def run_local_trigger_queue(
                 )
             except AutomationTurnError as exc:
                 error = str(exc) or exc.__class__.__name__
-                store.record_delivery(
+                await _await_delivery_settlement(
-                    delivery.trigger_id,
+                    _settle_failed_delivery(store, delivery, error=error),
-                    status="error",
+                    store=store,
-                    error=error,
+                    delivery=delivery,
-                    run_at_ms=delivery.created_at_ms,
                 )
-                _write_delivery_run_record(
-                    store,
-                    delivery,
-                    status="error",
-                    error=error,
-                )
-                store.complete_delivery(delivery)
                 logger.warning(
                     "Trigger: delivery {} for {} reached the agent but failed: {}",
                     delivery.id,
@@ -102,18 +97,10 @@ async def run_local_trigger_queue(
                 )
             except Exception as exc:
                 error = str(exc) or exc.__class__.__name__
-                retried = store.retry_delivery(delivery, error)
+                retried = await _await_delivery_settlement(
-                _write_delivery_run_record(
+                    _settle_retryable_delivery(store, delivery, error=error),
-                    store,
+                    store=store,
-                    delivery,
+                    delivery=delivery,
-                    status="retrying" if retried else "error",
-                    error=error,
-                )
-                store.record_delivery(
-                    delivery.trigger_id,
-                    status="error",
-                    error=error,
-                    run_at_ms=delivery.created_at_ms,
                 )
                 logger.exception(
                     "Trigger: failed delivery {} for {}{}",
@@ -127,6 +114,10 @@ class _TerminalDeliveryError(RuntimeError):
     pass
 
 
+class _DeliverySettledOnCancellation(asyncio.CancelledError):
+    """Cancellation reported only after an already-submitted delivery is settled."""
+
+
 async def _deliver_delivery(
     store: LocalTriggerStore,
     delivery: TriggerDelivery,
@@ -134,7 +125,7 @@ async def _deliver_delivery(
     submit_turn: Callable[[InboundMessage], Awaitable[OutboundMessage | None]],
     is_channel_enabled: Callable[[str], bool],
 ) -> None:
-    trigger = store.get(delivery.trigger_id)
+    trigger = await asyncio.to_thread(store.get, delivery.trigger_id)
     if trigger is None:
         raise _TerminalDeliveryError("trigger not found")
     if not trigger.enabled:
@@ -142,7 +133,14 @@ async def _deliver_delivery(
     if not is_channel_enabled(trigger.channel):
         raise _TerminalDeliveryError(f"target channel is not enabled: {trigger.channel}")
 
-    store.write_delivery_run_record(delivery, trigger=trigger, status="processing")
+    await shield_and_drain(
+        asyncio.to_thread(
+            store.write_delivery_run_record,
+            delivery,
+            trigger=trigger,
+            status="processing",
+        )
+    )
     msg = InboundMessage(
         channel=trigger.channel,
         sender_id=trigger.sender_id,
@@ -151,22 +149,177 @@ async def _deliver_delivery(
         metadata=_delivery_metadata(trigger, delivery),
         session_key_override=trigger.session_key,
     )
+    try:
         response = await submit_turn(msg)
-    store.record_delivery(
+    except AutomationTurnAcceptedCancellation:
+        try:
+            await _await_delivery_settlement(
+                _settle_accepted_delivery(store, delivery, trigger=trigger),
+                store=store,
+                delivery=delivery,
+            )
+        except _DeliverySettledOnCancellation:
+            raise
+        except Exception:
+            logger.exception(
+                "Trigger: failed to persist accepted delivery {}; dropping retry",
+                delivery.id,
+            )
+            with suppress(Exception):
+                await shield_and_drain(asyncio.to_thread(store.complete_delivery, delivery))
+        raise _DeliverySettledOnCancellation from None
+
+    try:
+        await _await_delivery_settlement(
+            _settle_submitted_delivery(store, delivery, trigger=trigger, response=response),
+            store=store,
+            delivery=delivery,
+        )
+    except Exception:
+        logger.exception(
+            "Trigger: failed to persist status for submitted delivery {}; dropping retry",
+            delivery.id,
+        )
+        with suppress(Exception):
+            await shield_and_drain(asyncio.to_thread(store.complete_delivery, delivery))
+
+
+async def _await_delivery_settlement(
+    operation: Awaitable[_T],
+    *,
+    store: LocalTriggerStore,
+    delivery: TriggerDelivery,
+) -> _T:
+    settlement = asyncio.ensure_future(operation)
+    try:
+        return await asyncio.shield(settlement)
+    except asyncio.CancelledError:
+        while not settlement.done():
+            try:
+                await asyncio.shield(settlement)
+            except asyncio.CancelledError:
+                continue
+        try:
+            settlement.result()
+        except Exception:
+            logger.exception(
+                "Trigger: failed to settle delivery {} during cancellation",
+                delivery.id,
+            )
+            completion = asyncio.create_task(
+                shield_and_drain(asyncio.to_thread(store.complete_delivery, delivery))
+            )
+            while not completion.done():
+                try:
+                    await asyncio.shield(completion)
+                except asyncio.CancelledError:
+                    continue
+            with suppress(Exception):
+                completion.result()
+        raise _DeliverySettledOnCancellation from None
+
+
+async def _settle_failed_delivery(
+    store: LocalTriggerStore,
+    delivery: TriggerDelivery,
+    *,
+    error: str,
+) -> None:
+    await _write_delivery_run_record(
+        store,
+        delivery,
+        status="error",
+        error=error,
+    )
+    await shield_and_drain(asyncio.to_thread(store.complete_delivery, delivery))
+    # Publish the terminal status only after the durable delivery state is settled.
+    await shield_and_drain(
+        asyncio.to_thread(
+            store.record_delivery,
+            delivery.trigger_id,
+            status="error",
+            error=error,
+            run_at_ms=delivery.created_at_ms,
+        )
+    )
+
+
+async def _settle_retryable_delivery(
+    store: LocalTriggerStore,
+    delivery: TriggerDelivery,
+    *,
+    error: str,
+) -> bool:
+    retried = await shield_and_drain(asyncio.to_thread(store.retry_delivery, delivery, error))
+    await _write_delivery_run_record(
+        store,
+        delivery,
+        status="retrying" if retried else "error",
+        error=error,
+    )
+    await shield_and_drain(
+        asyncio.to_thread(
+            store.record_delivery,
+            delivery.trigger_id,
+            status="error",
+            error=error,
+            run_at_ms=delivery.created_at_ms,
+        )
+    )
+    return retried
+
+
+async def _settle_accepted_delivery(
+    store: LocalTriggerStore,
+    delivery: TriggerDelivery,
+    *,
+    trigger: LocalTrigger,
+) -> None:
+    """Commit an accepted delivery without claiming the agent turn completed."""
+    await _write_delivery_run_record(
+        store,
+        delivery,
+        trigger=trigger,
+        status="accepted",
+    )
+    await shield_and_drain(asyncio.to_thread(store.complete_delivery, delivery))
+    await shield_and_drain(
+        asyncio.to_thread(
+            store.record_delivery,
             trigger.id,
             status="ok",
             run_at_ms=delivery.created_at_ms,
         )
-    _write_delivery_run_record(
+    )
+
+
+async def _settle_submitted_delivery(
+    store: LocalTriggerStore,
+    delivery: TriggerDelivery,
+    *,
+    trigger: LocalTrigger,
+    response: OutboundMessage | None,
+) -> None:
+    await _write_delivery_run_record(
         store,
         delivery,
         trigger=trigger,
         status="ok",
         response=response.content if response else "",
     )
+    await shield_and_drain(asyncio.to_thread(store.complete_delivery, delivery))
+    # last_status is the externally visible commit marker for a settled delivery.
+    await shield_and_drain(
+        asyncio.to_thread(
+            store.record_delivery,
+            trigger.id,
+            status="ok",
+            run_at_ms=delivery.created_at_ms,
+        )
+    )
 
 
-def _write_delivery_run_record(
+async def _write_delivery_run_record(
     store: LocalTriggerStore,
     delivery: TriggerDelivery,
     *,
@@ -176,13 +329,16 @@ def _write_delivery_run_record(
     response: str | None = None,
 ) -> None:
     try:
-        store.write_delivery_run_record(
+        await shield_and_drain(
+            asyncio.to_thread(
+                store.write_delivery_run_record,
                 delivery,
                 trigger=trigger,
                 status=status,
                 error=error,
                 response=response,
             )
+        )
     except Exception:
         logger.exception(
             "Trigger: failed to write run record for delivery {}",

nanobot/triggers/local_store.py

@@ -24,6 +24,7 @@ _MAX_RUN_HISTORY = 20
 _MAX_DELIVERY_ATTEMPTS = 10
 _RUN_RECORD_TEXT_MAX_CHARS = 4000
 _PROCESSING_RECOVERY_ERROR = "delivery was recovered from interrupted processing"
+_FILE_LOCK_TIMEOUT_SECONDS = 5
 
 
 class TriggerStoreError(RuntimeError):
@@ -49,7 +50,10 @@ class LocalTriggerStore:
         self.processing_dir = self.root / "processing"
         self.failed_dir = self.root / "failed"
         self.runs_dir = self.root / "runs"
-        self._lock = FileLock(str(self.root / ".lock"))
+        self._lock = FileLock(
+            str(self.root / ".lock"),
+            timeout=_FILE_LOCK_TIMEOUT_SECONDS,
+        )
 
     def create(
         self,

nanobot/utils/cancellation.py

@@ -3,8 +3,48 @@
 from __future__ import annotations
 
 import asyncio
+from collections.abc import Awaitable
+from typing import TypeVar
+
+_T = TypeVar("_T")
 
 
 def task_is_cancelling() -> bool:
     task = asyncio.current_task()
     return task is not None and task.cancelling() > 0
+
+
+async def shield_and_drain(awaitable: Awaitable[_T]) -> _T:
+    """Delay caller cancellation until an accepted operation has fully settled.
+
+    ``asyncio.to_thread`` cannot stop a worker that has already started. Shielding
+    keeps cancellation from detaching that worker, and draining also lets any
+    post-write in-memory settlement in ``awaitable`` finish. Cancellation is still
+    re-raised as soon as the accepted operation is done.
+    """
+    settlement = asyncio.ensure_future(awaitable)
+    cancellation: asyncio.CancelledError | None = None
+
+    while not settlement.done():
+        try:
+            result = await asyncio.shield(settlement)
+        except asyncio.CancelledError as exc:
+            if cancellation is None:
+                cancellation = exc
+        except BaseException:
+            if cancellation is None:
+                raise
+            break
+        else:
+            if cancellation is not None:
+                raise cancellation
+            return result
+
+    if cancellation is not None:
+        try:
+            settlement.result()
+        except BaseException:
+            # The caller's cancellation wins once settlement has been observed.
+            pass
+        raise cancellation
+    return settlement.result()

nanobot/webui/forking.py

@@ -2,6 +2,7 @@
 
 from __future__ import annotations
 
+import asyncio
 import re
 import uuid
 from collections.abc import Mapping
@@ -9,6 +10,7 @@ from typing import TYPE_CHECKING, Any, TypeGuard
 
 from nanobot.session.manager import SessionManager
 from nanobot.session.webui_turns import WEBUI_TITLE_METADATA_KEY, clean_generated_title
+from nanobot.utils.cancellation import shield_and_drain
 from nanobot.webui.transcript import (
     append_fork_marker,
     delete_webui_transcript,
@@ -93,15 +95,24 @@ async def handle_webui_fork_chat(
         await channel.send_webui_protocol_error(connection, "session_manager_unavailable")
         return
 
+    async def create_and_attach() -> None:
         try:
-        forked = create_webui_chat_fork(
+            forked = await asyncio.to_thread(
+                create_webui_chat_fork,
                 session_manager,
                 source_chat_id=source_chat_id,
                 before_user_index=raw_index,
-            title=envelope.get("title") if isinstance(envelope.get("title"), str) else None,
+                title=(
+                    envelope.get("title")
+                    if isinstance(envelope.get("title"), str)
+                    else None
+                ),
             )
             if forked is None:
-            await channel.send_webui_protocol_error(connection, "invalid fork source or index")
+                await channel.send_webui_protocol_error(
+                    connection,
+                    "invalid fork source or index",
+                )
                 return
             fork_id, fork_key = forked
         except Exception as exc:
@@ -114,3 +125,5 @@ async def handle_webui_fork_chat(
             fork_id=fork_id,
             fork_key=fork_key,
         )
+
+    await shield_and_drain(create_and_attach())

nanobot/webui/mcp_oauth_api.py

@@ -16,7 +16,7 @@ from nanobot.agent.tools.mcp import MCPConnection, connect_mcp_servers
 from nanobot.agent.tools.mcp_oauth import MCP_OAUTH_CALLBACK_PATH, MCPOAuthHandlers
 from nanobot.agent.tools.registry import ToolRegistry
 from nanobot.config.schema import MCPServerConfig
-from nanobot.security.network import validate_url_target
+from nanobot.security.network import async_validate_url_target
 from nanobot.webui.http_utils import is_loopback_host
 
 McpReload = Callable[[], Awaitable[dict[str, Any]]]
@@ -259,7 +259,7 @@ class McpOAuthManager:
         ):
             flow.error = "The MCP server returned an unsafe authorization URL."
             raise McpOAuthError(flow.error)
-        ok, _error = validate_url_target(authorization_url)
+        ok, _error = await async_validate_url_target(authorization_url)
         if not ok:
             flow.error = "The MCP server returned an unsafe authorization URL."
             raise McpOAuthError(flow.error)

nanobot/webui/settings_capabilities.py

@@ -27,6 +27,7 @@ from nanobot.providers.image_generation import (
 )
 from nanobot.providers.registry import find_by_name
 from nanobot.security.network import is_loopback_host
+from nanobot.utils.cancellation import shield_and_drain
 from nanobot.webui.settings_contracts import (
     QueryParams,
     SettingsRequest,
@@ -640,7 +641,11 @@ class CapabilitySettingsHandler:
     ) -> SettingsRouteResult:
         if action == "api-status":
             return SettingsRouteResult.success(
-                api_service_payload(self.settings, operations.api_runtime())
+                await asyncio.to_thread(
+                    api_service_payload,
+                    self.settings,
+                    operations.api_runtime(),
+                )
             )
         if action == "api-start":
             return await self._start_api(request, operations)
@@ -673,17 +678,22 @@ class CapabilitySettingsHandler:
             return SettingsRouteResult.failure(404, "unknown settings action")
 
         operation, section, apply_image_reload = mutation
-        try:
+
-            payload = self.settings.mutate(operation, request.query)
+        async def mutate_and_apply() -> tuple[dict[str, Any], bool]:
-        except WebUISettingsError as exc:
+            payload = await self.settings.mutate_async(operation, request.query)
-            return SettingsRouteResult.failure(exc.status, exc.message)
+            if not apply_image_reload:
-        if apply_image_reload:
+                return payload, False
-            payload, image_restart_cleared = await self.apply_image_runtime_change(
+            return await self.apply_image_runtime_change(
                 payload,
                 operations.reload_image,
             )
-        else:
+
-            image_restart_cleared = False
+        try:
+            payload, image_restart_cleared = await shield_and_drain(
+                mutate_and_apply()
+            )
+        except WebUISettingsError as exc:
+            return SettingsRouteResult.failure(exc.status, exc.message)
         return SettingsRouteResult.success(
             payload,
             decorate_restart=True,
@@ -726,16 +736,17 @@ class CapabilitySettingsHandler:
                 400,
                 "API service API key must be a string",
             )
-        try:
+        allow_install = await self._allow_feature_package_install(request)
-            await asyncio.to_thread(
+
-                self.settings.mutate,
+        async def mutate_and_start() -> Any:
+            await self.settings.mutate_async(
                 operations.nanobot_features_action,
                 "enable",
                 {"name": ["api"]},
-                allow_install=self._allow_feature_package_install(request),
+                allow_install=allow_install,
             )
-            self.settings.mutate(operations.update_api, request.query)
+            await self.settings.mutate_async(operations.update_api, request.query)
-            config = self.settings.config.load()
+            config = await self.settings.config.load_async()
             runtime = operations.api_runtime()
             options = ApiStartOptions(
                 host=config.api.host,
@@ -744,10 +755,13 @@ class CapabilitySettingsHandler:
                 config_path=str(self.settings.config.path),
             )
             current = runtime.status()
-            result = await asyncio.to_thread(
+            return await asyncio.to_thread(
                 runtime.restart if current.running else runtime.start_background,
                 options,
             )
+
+        try:
+            result = await shield_and_drain(mutate_and_start())
             if not result.ok:
                 return SettingsRouteResult.failure(
                     500,
@@ -762,7 +776,8 @@ class CapabilitySettingsHandler:
             self.logger.exception("failed to start managed API service")
             return SettingsRouteResult.failure(500, str(exc))
         return SettingsRouteResult.success(
-            api_service_payload(
+            await asyncio.to_thread(
+                api_service_payload,
                 self.settings,
                 operations.api_runtime(),
                 last_action="started",
@@ -775,7 +790,7 @@ class CapabilitySettingsHandler:
     ) -> SettingsRouteResult:
         runtime = operations.api_runtime()
         try:
-            result = await asyncio.to_thread(runtime.stop)
+            result = await shield_and_drain(asyncio.to_thread(runtime.stop))
         except Exception as exc:
             self.logger.exception("failed to stop managed API service")
             return SettingsRouteResult.failure(500, str(exc))
@@ -785,20 +800,20 @@ class CapabilitySettingsHandler:
                 api_runtime_message(result.message),
             )
         return SettingsRouteResult.success(
-            api_service_payload(
+            await asyncio.to_thread(
+                api_service_payload,
                 self.settings,
                 operations.api_runtime(),
                 last_action="stopped",
             )
         )
 
-    def _allow_feature_package_install(self, request: SettingsRequest) -> bool:
+    async def _allow_feature_package_install(self, request: SettingsRequest) -> bool:
         if request.local_browser:
             return True
         try:
-            return bool(
+            config = await self.settings.config.load_async()
-                self.settings.config.load().tools.webui_allow_remote_package_install
+            return bool(config.tools.webui_allow_remote_package_install)
-            )
         except Exception:
             self.logger.exception("failed to load remote package install policy")
             return False

nanobot/webui/settings_models.py

@@ -29,6 +29,7 @@ from nanobot.config.schema import Config, FallbackCandidate, ModelPresetConfig,
 from nanobot.providers.image_generation import get_image_gen_provider
 from nanobot.providers.oauth_guidance import OAUTH_CLI_KIT_MISSING_MESSAGE
 from nanobot.providers.registry import PROVIDERS, create_dynamic_spec, find_by_name
+from nanobot.utils.cancellation import shield_and_drain
 from nanobot.webui.settings_contracts import (
     QueryParams,
     SettingsRequest,
@@ -1651,6 +1652,30 @@ class ModelSettingsHandler:
         if self.settings.refresh_runtime_config is not None:
             self.settings.refresh_runtime_config()
 
+    async def _mutate_and_refresh(
+        self,
+        operation: SettingsOperation,
+        query: QueryParams,
+        **kwargs: Any,
+    ) -> dict[str, Any]:
+        payload = await self.settings.mutate_async(operation, query, **kwargs)
+        self._refresh_runtime_config()
+        return payload
+
+    async def _update_provider_and_runtime(
+        self,
+        operation: SettingsOperation,
+        query: QueryParams,
+        apply_image_runtime_change: Callable[
+            [dict[str, Any]],
+            Awaitable[tuple[dict[str, Any], bool]],
+        ],
+    ) -> tuple[dict[str, Any], bool]:
+        payload = await self.settings.mutate_async(operation, query)
+        payload, image_restart_cleared = await apply_image_runtime_change(payload)
+        self._refresh_runtime_config()
+        return payload, image_restart_cleared
+
     async def handle(
         self,
         action: str,
@@ -1659,8 +1684,12 @@ class ModelSettingsHandler:
     ) -> SettingsRouteResult:
         try:
             if action == "agent-update":
-                payload = self.settings.mutate(operations.update_agent, request.query)
+                payload = await shield_and_drain(
-                self._refresh_runtime_config()
+                    self._mutate_and_refresh(
+                        operations.update_agent,
+                        request.query,
+                    )
+                )
                 return SettingsRouteResult.success(
                     payload,
                     decorate_restart=True,
@@ -1668,12 +1697,13 @@ class ModelSettingsHandler:
                 )
 
             if action == "model-update":
-                payload = self.settings.mutate(
+                payload = await shield_and_drain(
+                    self._mutate_and_refresh(
                         operations.update_model,
                         request.query,
                         rename_model_preset=self.settings.rename_model_preset,
                     )
-                self._refresh_runtime_config()
+                )
                 return SettingsRouteResult.success(payload, decorate_restart=True)
 
             mutation = {
@@ -1684,19 +1714,19 @@ class ModelSettingsHandler:
                 "provider-create": operations.create_provider,
             }.get(action)
             if mutation is not None:
-                payload = self.settings.mutate(mutation, request.query)
+                payload = await shield_and_drain(
-                self._refresh_runtime_config()
+                    self._mutate_and_refresh(mutation, request.query)
+                )
                 return SettingsRouteResult.success(payload, decorate_restart=True)
 
             if action == "provider-update":
-                payload = self.settings.mutate(
+                payload, image_restart_cleared = await shield_and_drain(
+                    self._update_provider_and_runtime(
                         operations.update_provider,
                         request.query,
+                        operations.apply_image_runtime_change,
                     )
-                payload, image_restart_cleared = await operations.apply_image_runtime_change(
-                    payload
                 )
-                self._refresh_runtime_config()
                 return SettingsRouteResult.success(
                     payload,
                     decorate_restart=True,
@@ -1724,32 +1754,38 @@ class ModelSettingsHandler:
                 return SettingsRouteResult.success(payload)
 
             if action == "oauth-login":
-                payload = await asyncio.to_thread(
+                payload = await shield_and_drain(
+                    asyncio.to_thread(
                         self.settings.read,
                         operations.oauth_login,
                         request.query,
                         oauth_flows=self.settings.oauth_flows,
                     )
+                )
             elif action == "oauth-complete":
                 raw_response = (request.payload or {}).get("authorization_response")
                 if raw_response is not None and not isinstance(raw_response, str):
                     raise WebUISettingsError(
                         "OAuth authorization response must be a string"
                     )
-                payload = await asyncio.to_thread(
+                payload = await shield_and_drain(
+                    asyncio.to_thread(
                         self.settings.read,
                         operations.oauth_complete,
                         request.query,
                         raw_response or None,
                         oauth_flows=self.settings.oauth_flows,
                     )
+                )
             elif action == "oauth-logout":
-                payload = await asyncio.to_thread(
+                payload = await shield_and_drain(
+                    asyncio.to_thread(
                         self.settings.read,
                         operations.oauth_logout,
                         request.query,
                         oauth_flows=self.settings.oauth_flows,
                     )
+                )
             else:
                 return SettingsRouteResult.failure(404, "unknown settings action")
         except WebUISettingsError as exc:

nanobot/webui/settings_routes.py

@@ -4,6 +4,7 @@ from __future__ import annotations
 
 import asyncio
 import html
+import inspect
 import json
 from collections.abc import Awaitable, Callable, Mapping
 from typing import Any, cast
@@ -18,6 +19,7 @@ from nanobot.bus.queue import MessageBus
 from nanobot.channels.registry import load_channel_plugin
 from nanobot.channels.validation import validate_channel_config
 from nanobot.pairing import approve_code, deny_code, list_pending
+from nanobot.utils.cancellation import shield_and_drain
 from nanobot.webui import settings_capabilities as capability_domain
 from nanobot.webui import settings_contracts as contracts
 from nanobot.webui import settings_models as model_domain
@@ -208,6 +210,16 @@ def _payload_query(payload: dict[str, Any]) -> QueryParams:
     }
 
 
+async def _call_settings_handler(
+    handler: Callable[[], Response | Awaitable[Response]],
+) -> Response:
+    """Keep synchronous handlers off-loop while supporting native async handlers."""
+    result = await asyncio.to_thread(handler)
+    if inspect.isawaitable(result):
+        return await result
+    return result
+
+
 class WebUISettingsRouter:
     """Authenticate and dispatch settings requests to transport-neutral domains."""
 
@@ -284,9 +296,9 @@ class WebUISettingsRouter:
         if not self._authorized(request):
             return self._unauthorized()
         if route == ("root", "settings"):
-            return await asyncio.to_thread(self._handle_settings)
+            return await _call_settings_handler(self._handle_settings)
         if route == ("root", "usage"):
-            return await asyncio.to_thread(self._handle_settings_usage)
+            return await _call_settings_handler(self._handle_settings_usage)
 
         domain, action = route
         domain_request = self._domain_request(
@@ -415,19 +427,18 @@ class WebUISettingsRouter:
                 )
         return self._json_response(payload)
 
-    def _handle_settings(self) -> Response:
+    async def _handle_settings(self) -> Response:
-        return self._json_response(
+        payload = await self.settings.read_async(
-            self._with_restart_state(
-                self.settings.read(
             settings_payload,
             surface=self._runtime_surface,
             runtime_capability_overrides=self._runtime_capabilities,
         )
-            )
+        return self._json_response(self._with_restart_state(payload))
-        )
 
-    def _handle_settings_usage(self) -> Response:
+    async def _handle_settings_usage(self) -> Response:
-        return self._json_response(self.settings.read(settings_usage_payload))
+        return self._json_response(
+            await self.settings.read_async(settings_usage_payload)
+        )
 
     def _model_operations(self) -> model_domain.ModelSettingsOperations:
         return model_domain.ModelSettingsOperations(
@@ -560,7 +571,7 @@ class WebUISettingsRouter:
             allow_install=allow_install,
         )
 
-    def _allow_feature_package_install(
+    async def _allow_feature_package_install(
         self,
         connection: Any,
         request: WsRequest,
@@ -570,29 +581,33 @@ class WebUISettingsRouter:
             request,
             needs_local_browser=True,
         )
-        return self._system.allow_feature_package_install(domain_request)
+        return await self._system.allow_feature_package_install(domain_request)
 
     async def _handle_mcp_oauth_start(self, request: WsRequest) -> Response:
         if not self._authorized(request):
             return self._unauthorized()
-        if self._mcp_oauth_redirect_uri is None:
+        redirect_uri_for_request = self._mcp_oauth_redirect_uri
+        if redirect_uri_for_request is None:
             return self._error_response(500, "MCP OAuth callback is not configured")
         query = self._parse_mcp_settings_query(request)
-        try:
+
-            name, cfg = await asyncio.to_thread(
+        async def mutate_and_start() -> dict[str, Any]:
-                self.settings.mutate,
+            name, cfg = await self.settings.mutate_async(
                 ensure_mcp_oauth_server,
                 query,
             )
-            redirect_uri = self._mcp_oauth_redirect_uri(request)
+            redirect_uri = redirect_uri_for_request(request)
             reset = (_query_first(query, "reset") or "").lower() in {"1", "true", "yes"}
-            payload = await self._mcp_oauth.start(
+            return await self._mcp_oauth.start(
                 name,
                 cfg,
                 redirect_uri,
                 reload_mcp=self._reload_mcp_runtime,
                 reset_credentials=reset,
             )
+
+        try:
+            payload = await shield_and_drain(mutate_and_start())
         except Exception as exc:
             return self._mcp_oauth_error_response(exc, action="start")
         return self._json_response(payload)

nanobot/webui/settings_services.py

@@ -2,6 +2,7 @@
 
 from __future__ import annotations
 
+import asyncio
 import threading
 from collections.abc import Callable
 from dataclasses import dataclass
@@ -12,9 +13,11 @@ from filelock import FileLock
 
 from nanobot.config.loader import load_config, save_config
 from nanobot.config.schema import Config
+from nanobot.utils.cancellation import shield_and_drain
 
 _T = TypeVar("_T")
 _WEBUI_OAUTH_MAX_FLOWS = 8
+_SETTINGS_FILE_LOCK_TIMEOUT_SECONDS = 5
 
 
 class WebUISettingsConfig:
@@ -25,13 +28,20 @@ class WebUISettingsConfig:
         self.path.parent.mkdir(parents=True, exist_ok=True)
         self._lock = threading.RLock()
         lock_path = self.path.with_suffix(f"{self.path.suffix}.lock")
-        self._file_lock = FileLock(str(lock_path))
+        self._file_lock = FileLock(
+            str(lock_path),
+            timeout=_SETTINGS_FILE_LOCK_TIMEOUT_SECONDS,
+        )
 
     def load(self) -> Config:
         """Load this gateway's config without consulting the process-global path."""
         with self._lock:
             return load_config(self.path)
 
+    async def load_async(self) -> Config:
+        """Load config without running file I/O or lock waits on the event loop."""
+        return await asyncio.to_thread(self.load)
+
     def update(self, mutation: Callable[[Config], _T]) -> _T:
         """Apply and atomically persist one path-scoped read-modify-write operation."""
         with self._lock, self._file_lock:
@@ -40,11 +50,21 @@ class WebUISettingsConfig:
             save_config(config, self.path)
             return result
 
+    async def update_async(self, mutation: Callable[[Config], _T]) -> _T:
+        """Update config without blocking the event loop."""
+        return await shield_and_drain(asyncio.to_thread(self.update, mutation))
+
     def run_serialized(self, operation: Callable[[Path], _T]) -> _T:
         """Run a path-aware read-modify-write operation under the config-file lock."""
         with self._lock, self._file_lock:
             return operation(self.path)
 
+    async def run_serialized_async(self, operation: Callable[[Path], _T]) -> _T:
+        """Run a serialized config operation without blocking the event loop."""
+        return await shield_and_drain(
+            asyncio.to_thread(self.run_serialized, operation)
+        )
+
 
 class WebUIOAuthFlowRegistry:
     """Bounded, thread-safe OAuth flows owned by one gateway instance."""
@@ -146,6 +166,16 @@ class WebUISettingsServices:
         """Run a settings read against this gateway's explicit config path."""
         return operation(*args, config_path=self.config.path, **kwargs)
 
+    async def read_async(
+        self,
+        operation: Callable[..., _T],
+        /,
+        *args: Any,
+        **kwargs: Any,
+    ) -> _T:
+        """Run a settings read without blocking the event loop."""
+        return await asyncio.to_thread(self.read, operation, *args, **kwargs)
+
     def mutate(
         self,
         operation: Callable[..., _T],
@@ -161,3 +191,15 @@ class WebUISettingsServices:
                 **kwargs,
             )
         )
+
+    async def mutate_async(
+        self,
+        operation: Callable[..., _T],
+        /,
+        *args: Any,
+        **kwargs: Any,
+    ) -> _T:
+        """Mutate settings without blocking the event loop."""
+        return await shield_and_drain(
+            asyncio.to_thread(self.mutate, operation, *args, **kwargs)
+        )

nanobot/webui/settings_system.py

@@ -23,6 +23,7 @@ from nanobot.config.schema import Config
 from nanobot.llm_usage import llm_usage_payload
 from nanobot.optional_features import OptionalFeatureError, with_channel_runtime_status
 from nanobot.security.workspace_access import workspace_sandbox_status
+from nanobot.utils.cancellation import shield_and_drain
 from nanobot.webui.settings_capabilities import network_safety_payload
 from nanobot.webui.settings_contracts import (
     QueryParams,
@@ -446,12 +447,17 @@ class SystemSettingsHandler:
         operations: SystemSettingsOperations,
     ) -> SettingsRouteResult:
         try:
-            payload = await asyncio.to_thread(
+            pending = asyncio.to_thread(
                 operations.cli_apps_action,
                 action,
                 request.query,
                 config_path=self.settings.config.path,
             )
+            payload = (
+                await shield_and_drain(pending)
+                if action in {"install", "update", "uninstall"}
+                else await pending
+            )
         except WebUISettingsError as exc:
             return SettingsRouteResult.failure(exc.status, exc.message)
         except Exception as exc:
@@ -505,17 +511,29 @@ class SystemSettingsHandler:
         action: str,
         operations: SystemSettingsOperations,
     ) -> SettingsRouteResult:
-        try:
+        allow_install = (
+            action != "enable"
+            or await self.allow_feature_package_install(request)
+        )
+
+        async def mutate_and_apply() -> dict[str, Any]:
             payload = await asyncio.to_thread(
                 self._nanobot_features_action,
                 action,
                 request.query,
                 operations,
-                allow_install=(
+                allow_install=allow_install,
-                    action != "enable"
-                    or self.allow_feature_package_install(request)
-                ),
             )
+            payload = await self._apply_feature_runtime_change(
+                action,
+                request.query,
+                payload,
+                operations,
+            )
+            return self._with_channel_runtime_status(payload, operations)
+
+        try:
+            payload = await shield_and_drain(mutate_and_apply())
         except OptionalFeatureError as exc:
             return SettingsRouteResult.failure(exc.status, exc.message)
         except Exception as exc:
@@ -527,13 +545,6 @@ class SystemSettingsHandler:
                     action,
                 )
             return SettingsRouteResult.failure(status, message)
-        payload = await self._apply_feature_runtime_change(
-            action,
-            request.query,
-            payload,
-            operations,
-        )
-        payload = self._with_channel_runtime_status(payload, operations)
         return SettingsRouteResult.success(
             payload,
             decorate_restart=True,
@@ -628,6 +639,15 @@ class SystemSettingsHandler:
         self,
         request: SettingsRequest,
         operations: SystemSettingsOperations,
+    ) -> SettingsRouteResult:
+        return await shield_and_drain(
+            self._channel_configure_settled(request, operations)
+        )
+
+    async def _channel_configure_settled(
+        self,
+        request: SettingsRequest,
+        operations: SystemSettingsOperations,
     ) -> SettingsRouteResult:
         name = (query_first(request.query, "name") or "").strip()
         instance_id = (
@@ -682,7 +702,7 @@ class SystemSettingsHandler:
                 "enable",
                 feature_query,
                 operations,
-                allow_install=self.allow_feature_package_install(request),
+                allow_install=await self.allow_feature_package_install(request),
             )
         except OptionalFeatureError as exc:
             return SettingsRouteResult.failure(
@@ -825,6 +845,22 @@ class SystemSettingsHandler:
         channel_name: str,
         payload: dict[str, Any],
         operations: SystemSettingsOperations,
+    ) -> dict[str, Any]:
+        return await shield_and_drain(
+            self._settle_channel_connect_success(
+                request,
+                channel_name,
+                payload,
+                operations,
+            )
+        )
+
+    async def _settle_channel_connect_success(
+        self,
+        request: SettingsRequest,
+        channel_name: str,
+        payload: dict[str, Any],
+        operations: SystemSettingsOperations,
     ) -> dict[str, Any]:
         target = {"name": [channel_name]}
         if payload.get("instance_id"):
@@ -835,11 +871,11 @@ class SystemSettingsHandler:
                 "enable",
                 target,
                 operations,
-                allow_install=self.allow_feature_package_install(request),
+                allow_install=await self.allow_feature_package_install(request),
             )
         except OptionalFeatureError as exc:
             features = self.feature_runtime_fallback(
-                self._nanobot_features_payload(operations),
+                await asyncio.to_thread(self._nanobot_features_payload, operations),
                 message=(
                     f"{channel_name} connected, but enabling channel support failed: "
                     f"{exc.message}"
@@ -859,13 +895,12 @@ class SystemSettingsHandler:
         )
         return updated
 
-    def allow_feature_package_install(self, request: SettingsRequest) -> bool:
+    async def allow_feature_package_install(self, request: SettingsRequest) -> bool:
         if request.local_browser:
             return True
         try:
-            return bool(
+            config = await self.settings.config.load_async()
-                self.settings.config.load().tools.webui_allow_remote_package_install
+            return bool(config.tools.webui_allow_remote_package_install)
-            )
         except Exception:
             self.logger.exception("failed to load remote package install policy")
             return False
@@ -925,13 +960,18 @@ class SystemSettingsHandler:
         operations: SystemSettingsOperations,
     ) -> SettingsRouteResult:
         try:
-            payload = await operations.mcp_presets_action(
+            pending = operations.mcp_presets_action(
                 action,
                 request.query,
                 reload_mcp=operations.reload_mcp,
                 mcp_runtime_status=operations.mcp_runtime_status,
                 config=self.settings.config,
             )
+            payload = (
+                await pending
+                if action is None
+                else await shield_and_drain(pending)
+            )
         except Exception as exc:
             status = getattr(exc, "status", 500)
             message = getattr(exc, "message", str(exc))

nanobot/webui/ws_http.py

@@ -34,6 +34,7 @@ from nanobot.session.session_handles import (
     SessionHandleResolver,
 )
 from nanobot.triggers.local_types import LocalTrigger
+from nanobot.utils.cancellation import shield_and_drain
 from nanobot.webui.file_preview import (
     WebUIFilePreviewError,
     file_preview_availability_payload,
@@ -135,6 +136,18 @@ _WEBUI_MUTATION_PAYLOAD_ATTR = "_nanobot_webui_mutation_payload"
 _WEBUI_MUTATION_REQUEST_ATTR = "_nanobot_webui_mutation_request"
 _NO_STORE_HEADERS = [("Cache-Control", "no-store")]
 
+
+def _slow_http_operation(path: str) -> str:
+    """Return a route family without logging user-controlled path/query values."""
+    clean_path = path.split("?", 1)[0]
+    if clean_path == "/webui/bootstrap":
+        return clean_path
+    parts = [part for part in clean_path.split("/") if part]
+    if len(parts) >= 2 and parts[0] == "api":
+        return f"/api/{parts[1]}"
+    return "/webui"
+
+
 _WEBUI_MUTATION_PATHS = {
     "automation.enable": "/api/webui/automations/enable",
     "automation.disable": "/api/webui/automations/disable",
@@ -420,7 +433,12 @@ class GatewayHTTPHandler:
             response = await self._dispatch_resolved(connection, request, got)
             return response
         finally:
-            self._log_slow_http(got, response, started)
+            self._log_slow_http(
+                got,
+                response,
+                started,
+                input_chars=len(request.path),
+            )
 
     async def dispatch_webui_mutation(
         self,
@@ -556,7 +574,14 @@ class GatewayHTTPHandler:
 
         return connection.respond(404, "Not Found")
 
-    def _log_slow_http(self, path: str, response: Any | None, started: float) -> None:
+    def _log_slow_http(
+        self,
+        path: str,
+        response: Any | None,
+        started: float,
+        *,
+        input_chars: int,
+    ) -> None:
         elapsed_ms = int((time.perf_counter() - started) * 1000)
         if elapsed_ms < _SLOW_WEBUI_HTTP_LOG_MS:
             return
@@ -564,9 +589,10 @@ class GatewayHTTPHandler:
             return
         status = getattr(response, "status_code", None)
         self._log.warning(
-            "slow webui http route path={} status={} duration_ms={}",
+            "slow webui http operation={} status={} input_chars={} duration_ms={}",
-            path,
+            _slow_http_operation(path),
             status if status is not None else "none",
+            input_chars,
             elapsed_ms,
         )
 
@@ -694,7 +720,11 @@ class GatewayHTTPHandler:
     async def _dispatch_session_routes(self, request: WsRequest, got: str) -> Response | None:
         m = re.match(r"^/api/sessions/([^/]+)/webui-thread$", got)
         if m:
-            return self._handle_webui_thread_get(request, m.group(1))
+            return await asyncio.to_thread(
+                self._handle_webui_thread_get,
+                request,
+                m.group(1),
+            )
 
         m = re.match(r"^/api/sessions/([^/]+)/context$", got)
         if m:
@@ -702,15 +732,27 @@ class GatewayHTTPHandler:
 
         m = re.match(r"^/api/sessions/([^/]+)/file-preview$", got)
         if m:
-            return self._handle_file_preview(request, m.group(1))
+            return await asyncio.to_thread(
+                self._handle_file_preview,
+                request,
+                m.group(1),
+            )
 
         m = re.match(r"^/api/sessions/([^/]+)/automations$", got)
         if m:
-            return self._handle_session_automations(request, m.group(1))
+            return await self._run_cron_transaction(
+                self._handle_session_automations,
+                request,
+                m.group(1),
+            )
 
         m = re.match(r"^/api/sessions/([^/]+)/delete$", got)
         if m:
-            return self._handle_session_delete(request, m.group(1))
+            return await self._run_cron_transaction(
+                self._handle_session_delete,
+                request,
+                m.group(1),
+            )
 
         return None
 
@@ -957,13 +999,24 @@ class GatewayHTTPHandler:
 
     # -- Automation routes --------------------------------------------------
 
+    async def _run_cron_transaction(
+        self,
+        operation: Callable[..., Any],
+        /,
+        *args: Any,
+        **kwargs: Any,
+    ) -> Any:
+        if self.cron_service is not None:
+            return await self.cron_service.run_sync(operation, *args, **kwargs)
+        return await shield_and_drain(asyncio.to_thread(operation, *args, **kwargs))
+
     async def _dispatch_automation_routes(
         self,
         request: WsRequest,
         got: str,
     ) -> Response | None:
         if got == "/api/webui/automations":
-            return self._handle_webui_automations(request)
+            return await self._run_cron_transaction(self._handle_webui_automations, request)
         m = re.match(r"^/api/webui/automations/(enable|disable|delete|run|update)$", got)
         if m:
             return await self._handle_webui_automation_action(request, m.group(1))
@@ -1029,13 +1082,24 @@ class GatewayHTTPHandler:
         job_id = (_query_first(query, "id") or _query_first(query, "job_id") or "").strip()
         if not job_id:
             return _http_error(400, "missing automation id")
-        trigger = self.local_trigger_store.get(job_id) if self.local_trigger_store else None
+        trigger = (
+            await asyncio.to_thread(self.local_trigger_store.get, job_id)
+            if self.local_trigger_store
+            else None
+        )
         if trigger is not None:
-            return self._handle_local_trigger_action(request, action, trigger)
+            return await shield_and_drain(
+                asyncio.to_thread(
+                    self._handle_local_trigger_action,
+                    request,
+                    action,
+                    trigger,
+                )
+            )
 
         if self.cron_service is None:
             return _http_error(404, "automation not found")
-        job = self.cron_service.get_job(job_id)
+        job = await self.cron_service.run_sync(self.cron_service.get_job, job_id)
         if job is None:
             return _http_error(404, "automation not found")
         if job.payload.kind == "system_event":
@@ -1044,13 +1108,23 @@ class GatewayHTTPHandler:
             return _http_error(409, "automation has no linked chat")
 
         if action == "enable":
-            if self.cron_service.enable_job(job_id, enabled=True) is None:
+            result = await self.cron_service.run_sync(
+                self.cron_service.enable_job,
+                job_id,
+                enabled=True,
+            )
+            if result is None:
                 return _http_error(404, "automation not found")
         elif action == "disable":
-            if self.cron_service.enable_job(job_id, enabled=False) is None:
+            result = await self.cron_service.run_sync(
+                self.cron_service.enable_job,
+                job_id,
+                enabled=False,
+            )
+            if result is None:
                 return _http_error(404, "automation not found")
         elif action == "delete":
-            result = self.cron_service.remove_job(job_id)
+            result = await self.cron_service.run_sync(self.cron_service.remove_job, job_id)
             if result == "not_found":
                 return _http_error(404, "automation not found")
             if result == "protected":
@@ -1068,7 +1142,11 @@ class GatewayHTTPHandler:
             if isinstance(parsed, str):
                 return _http_error(400, parsed)
             try:
-                result = self.cron_service.update_job(job_id, **parsed)
+                result = await self.cron_service.run_sync(
+                    self.cron_service.update_job,
+                    job_id,
+                    **parsed,
+                )
             except ValueError as exc:
                 return _http_error(400, str(exc))
             if result == "not_found":
@@ -1078,7 +1156,7 @@ class GatewayHTTPHandler:
         else:
             return _http_error(404, "unknown automation action")
 
-        return self._handle_webui_automations(request)
+        return await self._run_cron_transaction(self._handle_webui_automations, request)
 
     def _handle_local_trigger_action(
         self,
@@ -1163,9 +1241,17 @@ class GatewayHTTPHandler:
         if got == "/api/webui/skills/install":
             return await self._handle_webui_skill_install(connection, request)
         if got == "/api/webui/skills/update":
-            return self._handle_webui_skill_update(request)
+            return await shield_and_drain(
+                asyncio.to_thread(self._handle_webui_skill_update, request)
+            )
         if got == "/api/webui/skills/delete":
-            return self._handle_webui_skill_delete(connection, request)
+            return await shield_and_drain(
+                asyncio.to_thread(
+                    self._handle_webui_skill_delete,
+                    connection,
+                    request,
+                )
+            )
         if got == "/api/webui/skills":
             return self._handle_webui_skills(request)
         m = re.match(r"^/api/webui/skills/([^/]+)$", got)
@@ -1276,7 +1362,7 @@ class GatewayHTTPHandler:
     ) -> Response:
         if not self.check_api_token(request):
             return _http_error(401, "Unauthorized")
-        if not self._allow_webui_package_install(connection, request):
+        if not await self._allow_webui_package_install(connection, request):
             return _http_error(403, "remote skill installation is disabled")
         if self._skill_install_lock.locked():
             return _http_error(409, "another skill installation is already in progress")
@@ -1308,13 +1394,16 @@ class GatewayHTTPHandler:
             "last_action": action,
         })
 
-    def _allow_webui_package_install(self, connection: Any, request: WsRequest) -> bool:
+    async def _allow_webui_package_install(
+        self,
+        connection: Any,
+        request: WsRequest,
+    ) -> bool:
         if _is_local_browser_request(connection, request.headers):
             return True
         try:
-            return bool(
+            config = await self.settings.config.load_async()
-                self.settings.config.load().tools.webui_allow_remote_package_install
+            return bool(config.tools.webui_allow_remote_package_install)
-            )
         except Exception:
             self._log.exception("failed to load remote package install policy")
             return False

tests/agent/test_mcp_reconnect_crash.py

@@ -123,21 +123,21 @@ def allow_loopback_mcp_urls(monkeypatch: pytest.MonkeyPatch):
         _resolver_lock = asyncio.Lock()
 
     monkeypatch.setattr(mcp_module, "PinnedDNSAsyncTransport", TestPinnedDNSAsyncTransport)
-    monkeypatch.setattr(
+    async def allow_url(url: str, *, allow_loopback: bool = False) -> tuple[bool, str]:
-        mcp_module,
+        return True, ""
-        "validate_url_target",
+
-        lambda url, *, allow_loopback=False: (True, ""),
+    async def resolve_url(
-    )
+        url: str,
-    monkeypatch.setattr(
+        *,
-        mcp_module,
+        allow_loopback: bool = False,
-        "resolve_url_target",
+        trust_remote_dns: bool = False,
-        lambda url, *, allow_loopback=False: (True, "", ("127.0.0.1",)),
+        timeout_s: float = 3.0,
-    )
+    ) -> tuple[bool, str, tuple[str, ...]]:
-    monkeypatch.setattr(
+        return True, "", ("127.0.0.1",)
-        security_network,
+
-        "resolve_url_target",
+    monkeypatch.setattr(mcp_module, "async_validate_url_target", allow_url)
-        lambda url, *, allow_loopback=False: (True, "", ("127.0.0.1",)),
+    monkeypatch.setattr(mcp_module, "async_resolve_url_target", resolve_url)
-    )
+    monkeypatch.setattr(security_network, "async_resolve_url_target", resolve_url)
     monkeypatch.setattr(
         mcp_module,
         "env_proxy_applies_to_url",

tests/agent/test_runner_errors.py

@@ -357,3 +357,50 @@ async def test_length_finish_with_blank_content_routes_to_length_recovery():
         "finish_reason='length' response with blank content"
     )
     assert result.final_content == "done"
+
+
+@pytest.mark.asyncio
+async def test_slow_tool_log_records_scale_without_argument_content(monkeypatch) -> None:
+    from nanobot.agent import runner as runner_module
+    from nanobot.agent.runner import AgentRunner
+
+    records: list[str] = []
+
+    class _Logger:
+        def warning(self, message: str, *args: object) -> None:
+            records.append(message.format(*args))
+
+    secret = "customer-token-do-not-log"
+
+    async def execute(_name, _args):
+        return "ok"
+
+    monkeypatch.setattr(runner_module, "_SLOW_TOOL_LOG_MS", 0)
+    monkeypatch.setattr(runner_module, "logger", _Logger())
+    runner = AgentRunner()
+    spec = make_run_spec(
+        MagicMock(spec=LLMProvider),
+        initial_messages=[],
+        tools=SimpleNamespace(execute=execute),
+        model="test-model",
+        max_iterations=1,
+        max_tool_result_chars=_MAX_TOOL_RESULT_CHARS,
+    )
+
+    await runner._run_tool(
+        spec,
+        ToolCallRequest(
+            id="call_1",
+            name="edit_file",
+            arguments={"old_text": secret, "paths": ["one", "two"]},
+        ),
+        external_lookup_counts={},
+        workspace_violation_counts={},
+    )
+
+    assert len(records) == 1
+    assert "operation=edit_file" in records[0]
+    assert "input_items=4" in records[0]
+    assert f"input_chars={len(secret)}" in records[0]
+    assert "duration_ms=" in records[0]
+    assert secret not in records[0]

tests/agent/test_runner_injections.py

@@ -1233,6 +1233,32 @@ async def test_submitted_cron_turn_reports_pending_until_completed(tmp_path):
     assert loop.pending_cron_job_ids_for_session(session_key) == set()
 
 
+@pytest.mark.asyncio
+async def test_accepted_automation_turn_preserves_wait_for_timeout(tmp_path):
+    """Accepted-turn signalling must not leak through asyncio timeout semantics."""
+    from nanobot.bus.events import InboundMessage
+    from nanobot.cron.session_turns import CRON_TRIGGER_META
+
+    loop = _make_loop(tmp_path)
+    loop._running = True
+    msg = InboundMessage(
+        channel="websocket",
+        sender_id="cron",
+        chat_id="chat-1",
+        content="scheduled work",
+        metadata={CRON_TRIGGER_META: {"job_id": "job-1", "run_id": "run-timeout"}},
+        session_key_override="websocket:chat-1",
+    )
+
+    submit_task = asyncio.create_task(
+        asyncio.wait_for(loop.submit_cron_turn(msg), timeout=0.01)
+    )
+    assert await asyncio.wait_for(loop.bus.consume_inbound(), timeout=0.5) is msg
+
+    with pytest.raises(TimeoutError):
+        await submit_task
+
+
 @pytest.mark.asyncio
 async def test_submitted_local_trigger_turn_reports_pending_until_completed(tmp_path):
     """Local triggers remain marked pending while their session turn is in flight."""
@@ -1818,3 +1844,53 @@ async def test_injection_cycle_cap_on_error_path():
     assert result.had_injections is True
     # Should cap: _MAX_INJECTION_CYCLES drained rounds + 1 final round that breaks
     assert call_count["n"] == _MAX_INJECTION_CYCLES + 1
+
+
+@pytest.mark.asyncio
+async def test_accepted_direct_automation_turn_keeps_single_effect_owner():
+    """Cancelling a submitter after direct admission must not cancel or replay its work."""
+    from nanobot.agent.automation_turns import AutomationTurnAcceptedCancellation
+    from nanobot.agent.cron_turns import CronTurnCoordinator
+    from nanobot.bus.events import InboundMessage
+    from nanobot.cron.session_turns import CRON_TRIGGER_META
+
+    dispatch_started = asyncio.Event()
+    release_dispatch = asyncio.Event()
+    effect_completed = asyncio.Event()
+    effects: list[str] = []
+
+    async def dispatch(msg: InboundMessage) -> None:
+        dispatch_started.set()
+        await release_dispatch.wait()
+        effects.append(msg.content)
+        coordinator.complete(msg)
+        effect_completed.set()
+
+    coordinator = CronTurnCoordinator(
+        publish_inbound=lambda _msg: asyncio.sleep(0),
+        dispatch=dispatch,
+        is_running=lambda: False,
+    )
+    msg = InboundMessage(
+        channel="websocket",
+        sender_id="cron",
+        chat_id="chat-1",
+        content="scheduled work",
+        metadata={CRON_TRIGGER_META: {"job_id": "job-1", "run_id": "run-direct"}},
+        session_key_override="websocket:chat-1",
+    )
+
+    submit_task = asyncio.create_task(coordinator.submit(msg))
+    try:
+        await asyncio.wait_for(dispatch_started.wait(), timeout=0.5)
+        submit_task.cancel()
+        with pytest.raises(AutomationTurnAcceptedCancellation):
+            await submit_task
+        assert effects == []
+
+        release_dispatch.set()
+        await asyncio.wait_for(effect_completed.wait(), timeout=0.5)
+        assert effects == ["scheduled work"]
+    finally:
+        release_dispatch.set()
+        await asyncio.gather(submit_task, return_exceptions=True)

tests/agent/test_workspace_scope.py

@@ -3,7 +3,6 @@ import os
 import subprocess
 import time
 from pathlib import Path
-from types import SimpleNamespace
 from unittest.mock import MagicMock
 
 import pytest
@@ -463,11 +462,21 @@ async def test_cli_app_scope_controls_working_dir(
 
     seen: dict[str, str] = {}
 
-    def fake_run(argv, **kwargs):
+    class _Process:
-        seen["cwd"] = kwargs["cwd"]
+        returncode = 0
-        return SimpleNamespace(returncode=0, stdout="ok", stderr="")
 
-    monkeypatch.setattr("nanobot.apps.cli.service.subprocess.run", fake_run)
+        async def communicate(self) -> tuple[bytes, bytes]:
+            return b"ok", b""
+
+    async def fake_create_subprocess_exec(*_argv: str, **kwargs: object) -> _Process:
+        seen["cwd"] = str(kwargs["cwd"])
+        return _Process()
+
+    monkeypatch.setattr(
+        "nanobot.apps.cli.service.asyncio.create_subprocess_exec",
+        fake_create_subprocess_exec,
+    )
+    monkeypatch.setattr("nanobot.apps.cli.service._WindowsJob.create", lambda: None)
     tool = CliAppsTool(
         workspace=tmp_path,
         restrict_to_workspace=True,

tests/agent/tools/test_long_task.py

@@ -3,6 +3,7 @@
 from __future__ import annotations
 
 import asyncio
+import threading
 from unittest.mock import AsyncMock, MagicMock
 
 import pytest
@@ -21,9 +22,9 @@ from nanobot.agent.tools.long_task import (
 from nanobot.agent.tools.registry import ToolRegistry
 from nanobot.bus.outbound_events import GoalStateSyncEvent
 from nanobot.bus.queue import MessageBus
-from nanobot.bus.runtime_events import RuntimeEventBus
+from nanobot.bus.runtime_events import GoalStateChanged, RuntimeEventBus
 from nanobot.session.goal_state import GOAL_STATE_KEY, MAX_GOAL_OBJECTIVE_CHARS
-from nanobot.session.manager import SessionManager
+from nanobot.session.manager import Session, SessionManager
 from nanobot.session.turn_continuation import should_finalize_on_max_iterations
 from nanobot.session.webui_turns import WebuiTurnCoordinator
 
@@ -179,13 +180,13 @@ async def test_goal_state_mutations_roll_back_on_save_failure(tmp_path, monkeypa
     sess = sm.get_or_create("websocket:c1")
     sess.metadata["marker"] = {"keep": True}
     sess.metadata["_sustained_goal_continuation_rounds"] = 12
-    original_save = sm.save
+    original_save_async = sm.save_async
     create_context = _request_context()
 
-    def fail_save(_session, **_kwargs):
+    async def fail_save(_session, **_kwargs):
         raise OSError("disk unavailable")
 
-    monkeypatch.setattr(sm, "save", fail_save)
+    monkeypatch.setattr(sm, "save_async", fail_save)
     with pytest.raises(OSError, match="disk unavailable"):
         await _execute(create, create_context, objective="Old")
 
@@ -195,13 +196,13 @@ async def test_goal_state_mutations_roll_back_on_save_failure(tmp_path, monkeypa
     }
     assert GOAL_STATE_KEY not in SessionManager(tmp_path).get_or_create("websocket:c1").metadata
 
-    monkeypatch.setattr(sm, "save", original_save)
+    monkeypatch.setattr(sm, "save_async", original_save_async)
     assert "Goal recorded" in await _execute(create, create_context, objective="Old")
     sess.metadata["_sustained_goal_continuation_rounds"] = 12
     sm.save(sess)
     replace_context = _request_context()
 
-    monkeypatch.setattr(sm, "save", fail_save)
+    monkeypatch.setattr(sm, "save_async", fail_save)
     with pytest.raises(OSError, match="disk unavailable"):
         await _execute(update, replace_context, action="replace", objective="New")
 
@@ -211,7 +212,7 @@ async def test_goal_state_mutations_roll_back_on_save_failure(tmp_path, monkeypa
     assert persisted[GOAL_STATE_KEY]["objective"] == "Old"
     assert persisted["_sustained_goal_continuation_rounds"] == 12
 
-    monkeypatch.setattr(sm, "save", original_save)
+    monkeypatch.setattr(sm, "save_async", original_save_async)
     assert "Goal replaced" in await _execute(
         update,
         replace_context,
@@ -225,6 +226,57 @@ async def test_goal_state_mutations_roll_back_on_save_failure(tmp_path, monkeypa
     )
 
 
+@pytest.mark.asyncio
+async def test_sync_goal_save_cancellation_settles_state_and_runtime_event() -> None:
+    started = threading.Event()
+    release = threading.Event()
+    session = Session(key="websocket:c1")
+    durable_metadata: dict[str, object] = {}
+    save_calls = 0
+
+    class _SyncSessionManager:
+        def get_or_create(self, key: str) -> Session:
+            assert key == session.key
+            return session
+
+        def save(self, target: Session) -> None:
+            nonlocal durable_metadata, save_calls
+            save_calls += 1
+            started.set()
+            assert release.wait(timeout=1)
+            durable_metadata = dict(target.metadata)
+
+    runtime_events = RuntimeEventBus()
+    published: list[GoalStateChanged] = []
+    runtime_events.subscribe(published.append, GoalStateChanged)
+    create = CreateGoalTool(
+        sessions=_SyncSessionManager(),  # type: ignore[arg-type]
+        runtime_events=runtime_events,
+    )
+    task = asyncio.create_task(
+        _execute(create, _request_context(), objective="Persist through cancellation")
+    )
+    assert await asyncio.to_thread(started.wait, 1)
+    try:
+        task.cancel()
+        await asyncio.sleep(0)
+        assert not task.done()
+    finally:
+        release.set()
+
+    with pytest.raises(asyncio.CancelledError):
+        await asyncio.wait_for(task, timeout=1)
+
+    assert save_calls == 1
+    assert session.metadata[GOAL_STATE_KEY]["objective"] == "Persist through cancellation"
+    assert durable_metadata[GOAL_STATE_KEY] == session.metadata[GOAL_STATE_KEY]
+    assert len(published) == 1
+    assert published[0].session_metadata[GOAL_STATE_KEY] == session.metadata[GOAL_STATE_KEY]
+    await asyncio.sleep(0.05)
+    assert save_calls == 1
+    assert len(published) == 1
+
+
 @pytest.mark.asyncio
 async def test_goal_tools_reject_oversized_objectives(tmp_path):
     sm = SessionManager(tmp_path)

tests/apps/test_cli_subprocess_env.py

@@ -42,17 +42,21 @@ def test_run_passes_filtered_env(monkeypatch, tmp_path) -> None:
     manager = CliAppManager(workspace=tmp_path, data_dir=tmp_path / "cli-apps")
     captured: dict[str, object] = {}
 
-    def fake_run(*args, **kwargs):
+    class _Process:
-        captured.update(kwargs)
-
-        class Result:
         returncode = 0
-            stdout = "ok"
-            stderr = ""
 
-        return Result()
+        def communicate(self, *, timeout: int) -> tuple[str, str]:
+            assert timeout == 60
+            return "ok", ""
 
-    monkeypatch.setattr("nanobot.apps.cli.service.subprocess.run", fake_run)
+    def fake_popen(argv: list[str], **kwargs: object) -> _Process:
+        assert argv == ["/bin/echo", "hi"]
+        assert kwargs["text"] is True
+        captured.update(kwargs)
+        return _Process()
+
+    monkeypatch.setattr("nanobot.apps.cli.service.subprocess.Popen", fake_popen)
+    monkeypatch.setattr("nanobot.apps.cli.service._WindowsJob.create", lambda: None)
     monkeypatch.setattr(manager, "get_app", lambda name: {"name": name, "entry_point": "echo"})
     monkeypatch.setattr(
         manager,

tests/cli/test_gateway_runtime.py

@@ -8,10 +8,15 @@ block the stop.
 """
 
 import asyncio
+import threading
 import time
 from contextlib import suppress
 
-from nanobot.cli.gateway_runtime import _close_gateway_runtime
+from nanobot.cli.gateway_runtime import (
+    _call_session_manager,
+    _close_gateway_runtime,
+    _monitor_event_loop_lag,
+)
 
 
 class _FakeAgent:
@@ -216,3 +221,110 @@ async def test_cancelled_runtime_tasks_gather_does_not_raise() -> None:
     assert runtime_tasks.done()  # the cancelled gather was awaited without raising
     assert agent.close_calls == 1
     assert provider.close_calls == 1
+
+
+async def test_session_manager_call_prefers_class_declared_coroutine() -> None:
+    class _Manager:
+        async def list_sessions_async(self) -> list[str]:
+            return ["native-async"]
+
+        def list_sessions(self) -> list[str]:
+            raise AssertionError("sync fallback should not run")
+
+    manager = _Manager()
+
+    result = await _call_session_manager(
+        manager,
+        "list_sessions_async",
+        manager.list_sessions,
+    )
+
+    assert result == ["native-async"]
+
+
+async def test_session_manager_call_offloads_sync_compatibility_fallback() -> None:
+    calling_thread = threading.get_ident()
+    sync_threads: list[int] = []
+
+    class _Manager:
+        def list_sessions(self) -> list[str]:
+            sync_threads.append(threading.get_ident())
+            return ["sync-fallback"]
+
+    manager = _Manager()
+
+    async def _fabricated_async() -> list[str]:
+        raise AssertionError("instance-only async stand-in should not run")
+
+    setattr(manager, "list_sessions_async", _fabricated_async)
+    result = await _call_session_manager(
+        manager,
+        "list_sessions_async",
+        manager.list_sessions,
+    )
+
+    assert result == ["sync-fallback"]
+    assert sync_threads and sync_threads[0] != calling_thread
+
+
+async def test_session_manager_sync_fallback_cancellation_waits_for_worker() -> None:
+    started = threading.Event()
+    release = threading.Event()
+    finished = threading.Event()
+    mutations: list[str] = []
+
+    class _Manager:
+        def save(self, value: str) -> str:
+            started.set()
+            assert release.wait(timeout=1)
+            mutations.append(value)
+            finished.set()
+            return "saved"
+
+    manager = _Manager()
+    task = asyncio.create_task(
+        _call_session_manager(manager, "save_async", manager.save, "mutation")
+    )
+    assert await asyncio.to_thread(started.wait, 1)
+    try:
+        task.cancel()
+        await asyncio.sleep(0)
+        assert not task.done()
+    finally:
+        release.set()
+
+    with suppress(asyncio.CancelledError):
+        await asyncio.wait_for(task, timeout=1)
+
+    assert task.cancelled()
+    assert finished.is_set()
+    assert mutations == ["mutation"]
+    await asyncio.sleep(0.05)
+    assert mutations == ["mutation"]
+
+
+async def test_event_loop_lag_monitor_logs_gateway_scheduler_drift() -> None:
+    records: list[str] = []
+
+    class _Logger:
+        def warning(self, message: str, *args: object) -> None:
+            records.append(message.format(*args))
+
+    task = asyncio.create_task(
+        _monitor_event_loop_lag(
+            interval_s=0.01,
+            warning_threshold_s=0.015,
+            log=_Logger(),
+        )
+    )
+    await asyncio.sleep(0)
+    time.sleep(0.04)
+    await asyncio.sleep(0.02)
+    task.cancel()
+    with suppress(asyncio.CancelledError):
+        await task
+
+    assert records
+    assert "operation=gateway" in records[0]
+    assert "duration_ms=" in records[0]
+    assert "interval_ms=10" in records[0]

tests/cli_apps/test_service.py

@@ -1,9 +1,11 @@
 from __future__ import annotations
 
+import asyncio
 import json
 import subprocess
 import sys
 import time
+from contextlib import suppress
 from pathlib import Path
 from types import SimpleNamespace
 
@@ -944,19 +946,23 @@ def test_run_installed_cli_uses_argv_without_shell(
         lambda entry: resolved if entry == "cli-anything-gimp" else None,
     )
 
-    def fake_run(argv: list[str], **kwargs: object) -> subprocess.CompletedProcess[str]:
+    class _Process:
-        assert "shell" not in kwargs or kwargs["shell"] is False
+        returncode = 0
+
+        def communicate(self, *, timeout: int) -> tuple[str, str]:
+            assert timeout == 5
+            return "ARGS=['--json', 'project', 'list']", ""
+
+    def fake_popen(argv: list[str], **kwargs: object) -> _Process:
+        assert argv == [resolved, "--json", "project", "list"]
+        assert "shell" not in kwargs
         assert kwargs["text"] is True
         assert kwargs["encoding"] == "utf-8"
         assert kwargs["errors"] == "replace"
-        return subprocess.CompletedProcess(
+        return _Process()
-            argv,
-            0,
-            stdout="ARGS=" + repr(argv[1:]),
-            stderr="",
-        )
 
-    monkeypatch.setattr("nanobot.apps.cli.service.subprocess.run", fake_run)
+    monkeypatch.setattr("nanobot.apps.cli.service.subprocess.Popen", fake_popen)
+    monkeypatch.setattr("nanobot.apps.cli.service._WindowsJob.create", lambda: None)
     manager._save_installed(
         {
             "gimp": {
@@ -986,12 +992,21 @@ def test_run_reports_created_artifacts(
         lambda entry: resolved if entry == "cli-anything-gimp" else None,
     )
 
-    def fake_run(argv: list[str], **kwargs: object) -> subprocess.CompletedProcess[str]:
+    class _Process:
+        returncode = 0
+
+        def communicate(self, *, timeout: int) -> tuple[str, str]:
+            assert timeout == 5
+            return "done", ""
+
+    def fake_popen(argv: list[str], **kwargs: object) -> _Process:
+        assert argv == [resolved, "render"]
         cwd = Path(str(kwargs["cwd"]))
         (cwd / "diagram.png").write_bytes(b"\x89PNG\r\n\x1a\nimage")
-        return subprocess.CompletedProcess(argv, 0, stdout="done", stderr="")
+        return _Process()
 
-    monkeypatch.setattr("nanobot.apps.cli.service.subprocess.run", fake_run)
+    monkeypatch.setattr("nanobot.apps.cli.service.subprocess.Popen", fake_popen)
+    monkeypatch.setattr("nanobot.apps.cli.service._WindowsJob.create", lambda: None)
     manager._save_installed({"gimp": {"entry_point": "cli-anything-gimp"}})
 
     result = manager.run("gimp", ["render"])
@@ -1100,3 +1115,144 @@ def test_uninstall_uses_uv_pip_when_pip_unavailable(
         sys.executable,
         "suno-cli",
     ]
+
+
+@pytest.mark.asyncio
+async def test_run_async_keeps_event_loop_responsive(
+    tmp_path: Path,
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    manager = _manager(tmp_path)
+    _seed_catalog(manager)
+    manager._save_installed({"gimp": {"entry_point": "python"}})
+    monkeypatch.setattr(
+        "nanobot.apps.cli.service.shutil.which",
+        lambda entry: sys.executable if entry == "python" else None,
+    )
+
+    task = asyncio.create_task(
+        manager.run_async(
+            "gimp",
+            ["-c", "import time; time.sleep(0.08); print('done')"],
+        )
+    )
+    for _ in range(3):
+        await asyncio.sleep(0.01)
+    assert not task.done()
+
+    result = await asyncio.wait_for(task, timeout=2)
+    assert "CLI app 'gimp' exited 0" in result
+    assert "done" in result
+
+
+@pytest.mark.asyncio
+async def test_run_async_cancellation_terminates_process_tree(
+    tmp_path: Path,
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    manager = _manager(tmp_path)
+    _seed_catalog(manager)
+    manager._save_installed({"gimp": {"entry_point": "python"}})
+    monkeypatch.setattr(
+        "nanobot.apps.cli.service.shutil.which",
+        lambda entry: sys.executable if entry == "python" else None,
+    )
+    started = manager.workspace / "parent-started.txt"
+    orphan_marker = manager.workspace / "orphan-survived.txt"
+    child_code = (
+        "import time; from pathlib import Path; time.sleep(0.5); "
+        f"Path({str(orphan_marker)!r}).write_text('orphan', encoding='utf-8')"
+    )
+    parent_code = (
+        "import subprocess, sys, time; from pathlib import Path; "
+        f"subprocess.Popen([sys.executable, '-c', {child_code!r}]); "
+        f"Path({str(started)!r}).write_text('started', encoding='utf-8'); "
+        "time.sleep(30)"
+    )
+    task = asyncio.create_task(manager.run_async("gimp", ["-c", parent_code]))
+    try:
+        for _ in range(200):
+            if started.exists():
+                break
+            await asyncio.sleep(0.01)
+        assert started.exists()
+
+        task.cancel()
+        with pytest.raises(asyncio.CancelledError):
+            await asyncio.wait_for(task, timeout=3)
+        await asyncio.sleep(0.8)
+        assert not orphan_marker.exists()
+    finally:
+        if not task.done():
+            task.cancel()
+            with suppress(asyncio.CancelledError):
+                await task
+
+
+def _parent_exits_with_pipe_holding_descendant(started: Path, marker: Path) -> str:
+    child_code = (
+        "import time; from pathlib import Path; time.sleep(1.5); "
+        f"Path({str(marker)!r}).write_text('survived', encoding='utf-8')"
+    )
+    return (
+        "import subprocess, sys, time; from pathlib import Path; "
+        "time.sleep(0.25); "
+        f"subprocess.Popen([sys.executable, '-c', {child_code!r}]); "
+        f"Path({str(started)!r}).write_text('spawned', encoding='utf-8')"
+    )
+
+
+@pytest.mark.asyncio
+async def test_run_async_timeout_kills_pipe_holding_descendant_after_parent_exit(
+    tmp_path: Path,
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    manager = _manager(tmp_path)
+    _seed_catalog(manager)
+    manager._save_installed({"gimp": {"entry_point": "python"}})
+    monkeypatch.setattr(
+        "nanobot.apps.cli.service.shutil.which",
+        lambda entry: sys.executable if entry == "python" else None,
+    )
+    started = manager.workspace / "descendant-started.txt"
+    marker = manager.workspace / "descendant-survived.txt"
+    parent_code = _parent_exits_with_pipe_holding_descendant(started, marker)
+
+    before = time.monotonic()
+    result = await asyncio.wait_for(
+        manager.run_async("gimp", ["-c", parent_code], timeout=1),
+        timeout=4,
+    )
+    elapsed = time.monotonic() - before
+
+    assert started.exists()
+    assert result == "CLI app 'gimp' timed out after 1s"
+    assert elapsed < 4
+    await asyncio.sleep(0.9)
+    assert not marker.exists()
+
+
+def test_run_timeout_kills_pipe_holding_descendant_after_parent_exit(
+    tmp_path: Path,
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    manager = _manager(tmp_path)
+    _seed_catalog(manager)
+    manager._save_installed({"gimp": {"entry_point": "python"}})
+    monkeypatch.setattr(
+        "nanobot.apps.cli.service.shutil.which",
+        lambda entry: sys.executable if entry == "python" else None,
+    )
+    started = manager.workspace / "sync-descendant-started.txt"
+    marker = manager.workspace / "sync-descendant-survived.txt"
+    parent_code = _parent_exits_with_pipe_holding_descendant(started, marker)
+
+    before = time.monotonic()
+    result = manager.run("gimp", ["-c", parent_code], timeout=1)
+    elapsed = time.monotonic() - before
+
+    assert started.exists()
+    assert result == "CLI app 'gimp' timed out after 1s"
+    assert elapsed < 4
+    time.sleep(0.9)
+    assert not marker.exists()

tests/cli_apps/test_tool.py

@@ -2,10 +2,13 @@ from __future__ import annotations
 
 import asyncio
 import json
-import subprocess
+import threading
 import time
 from pathlib import Path
 
+import pytest
+
+from nanobot.agent.tools import cli_apps as cli_apps_tool
 from nanobot.agent.tools.cli_apps import CliAppsTool
 from nanobot.agent.tools.context import RequestContext
 from nanobot.apps.cli.service import CliAppManager, CliAppsRuntimeConfig
@@ -52,16 +55,22 @@ def test_run_cli_app_uses_installed_registry_app(
         lambda entry: resolved if entry == "cli-anything-gimp" else None,
     )
 
-    def fake_run(argv: list[str], **kwargs: object) -> subprocess.CompletedProcess[str]:
+    class _Process:
-        assert "shell" not in kwargs or kwargs["shell"] is False
+        returncode = 0
-        return subprocess.CompletedProcess(
-            argv,
-            0,
-            stdout="tool:" + " ".join(argv[1:]),
-            stderr="",
-        )
 
-    monkeypatch.setattr("nanobot.apps.cli.service.subprocess.run", fake_run)
+        async def communicate(self) -> tuple[bytes, bytes]:
+            return b"tool:--json project list", b""
+
+    async def fake_create_subprocess_exec(*argv: str, **kwargs: object) -> _Process:
+        assert argv == (resolved, "--json", "project", "list")
+        assert "shell" not in kwargs
+        return _Process()
+
+    monkeypatch.setattr(
+        "nanobot.apps.cli.service.asyncio.create_subprocess_exec",
+        fake_create_subprocess_exec,
+    )
+    monkeypatch.setattr("nanobot.apps.cli.service._WindowsJob.create", lambda: None)
     monkeypatch.setattr("nanobot.apps.cli.service.get_runtime_subdir", lambda _name: data_dir)
 
     tool = CliAppsTool(
@@ -156,3 +165,90 @@ def test_cli_app_tool_provides_context_only_for_attachment(tmp_path: Path) -> No
     assert attached is not None
     assert attached.source == "cli_apps"
     assert "CLI App Attachment: @drawio" in attached.content
+
+
+@pytest.mark.asyncio
+async def test_run_cli_app_uses_threaded_sync_fallback_for_compatible_manager(
+    tmp_path: Path,
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    event_loop_thread = threading.get_ident()
+    seen: dict[str, object] = {}
+
+    class CompatibleManager:
+        def __init__(
+            self,
+            *,
+            workspace: Path,
+            runtime: CliAppsRuntimeConfig,
+        ) -> None:
+            seen["workspace"] = workspace
+            seen["runtime"] = runtime
+
+        def run_async(self, *_args: object, **_kwargs: object) -> str:
+            raise AssertionError("a synchronous compatibility method must not be awaited")
+
+        def run(self, name: str, **kwargs: object) -> str:
+            seen["thread"] = threading.get_ident()
+            seen["name"] = name
+            seen["kwargs"] = kwargs
+            return "compatible result"
+
+    monkeypatch.setattr(cli_apps_tool, "CliAppManager", CompatibleManager)
+    runtime = CliAppsRuntimeConfig(run_timeout=7)
+    tool = CliAppsTool(workspace=tmp_path, restrict_to_workspace=True, runtime=runtime)
+
+    result = await tool.execute(
+        name="demo",
+        args=["show"],
+        json=True,
+        working_dir=str(tmp_path),
+        timeout=3,
+    )
+
+    assert result == "compatible result"
+    assert seen["workspace"] == tmp_path
+    assert seen["runtime"] is runtime
+    assert seen["thread"] != event_loop_thread
+    assert seen["name"] == "demo"
+    assert seen["kwargs"] == {
+        "args": ["show"],
+        "json_output": True,
+        "working_dir": str(tmp_path),
+        "timeout": 3,
+        "restrict_to_workspace": True,
+    }
+
+
+@pytest.mark.asyncio
+async def test_run_cli_app_preserves_native_manager_cancellation(
+    tmp_path: Path,
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    started = asyncio.Event()
+    cancelled = asyncio.Event()
+
+    async def fake_run_async(self: CliAppManager, name: str, **_kwargs: object) -> str:
+        assert name == "demo"
+        started.set()
+        try:
+            await asyncio.Future()
+        except asyncio.CancelledError:
+            cancelled.set()
+            raise
+        raise AssertionError("unreachable")
+
+    def fail_sync_run(self: CliAppManager, name: str, **_kwargs: object) -> str:
+        raise AssertionError(f"native async manager fell back to run({name!r})")
+
+    monkeypatch.setattr(CliAppManager, "run_async", fake_run_async)
+    monkeypatch.setattr(CliAppManager, "run", fail_sync_run)
+    tool = CliAppsTool(workspace=tmp_path)
+
+    task = asyncio.create_task(tool.execute(name="demo"))
+    await asyncio.wait_for(started.wait(), timeout=1)
+    task.cancel()
+
+    with pytest.raises(asyncio.CancelledError):
+        await task
+    assert cancelled.is_set()

tests/command/test_router_dispatchable.py

@@ -2,6 +2,8 @@
 
 from __future__ import annotations
 
+import asyncio
+import threading
 from inspect import Parameter, signature
 from unittest.mock import AsyncMock, MagicMock
 
@@ -9,9 +11,11 @@ import pytest
 
 from nanobot.command.builtin import (
     builtin_command_starts_agent_turn,
+    cmd_new,
     register_builtin_commands,
 )
 from nanobot.command.router import CommandContext, CommandRouter
+from nanobot.session.manager import Session
 
 
 def test_command_context_requires_loop_as_keyword_dependency() -> None:
@@ -95,6 +99,55 @@ def test_builtin_command_agent_turn_lifecycle(content: str, expected: bool) -> N
     assert builtin_command_starts_agent_turn(content) is expected
 
 
+@pytest.mark.asyncio
+async def test_new_cancellation_waits_for_save_and_cache_invalidation() -> None:
+    started = threading.Event()
+    release = threading.Event()
+    session = Session(key="test:chat1")
+    invalidated: list[str] = []
+    save_calls = 0
+
+    class _SyncSessionManager:
+        def save(self, target: Session) -> None:
+            nonlocal save_calls
+            assert target is session
+            save_calls += 1
+            started.set()
+            assert release.wait(timeout=1)
+
+        def invalidate(self, key: str) -> None:
+            invalidated.append(key)
+
+    loop = MagicMock()
+    loop.sessions = _SyncSessionManager()
+    loop._cancel_active_tasks = AsyncMock(return_value=0)
+    ctx = CommandContext(
+        msg=MagicMock(channel="test", chat_id="chat1", metadata={}),
+        session=session,
+        key=session.key,
+        raw="/new",
+        loop=loop,
+    )
+    task = asyncio.create_task(cmd_new(ctx))
+    assert await asyncio.to_thread(started.wait, 1)
+    try:
+        task.cancel()
+        await asyncio.sleep(0)
+        assert not task.done()
+        assert invalidated == []
+    finally:
+        release.set()
+
+    with pytest.raises(asyncio.CancelledError):
+        await asyncio.wait_for(task, timeout=1)
+
+    assert save_calls == 1
+    assert invalidated == [session.key]
+    await asyncio.sleep(0.05)
+    assert save_calls == 1
+    assert invalidated == [session.key]
+
+
 class TestMidTurnCommandDispatchedDirectly:
     """Verify that commands matching is_dispatchable_command() are dispatched
     correctly when session=None (the mid-turn path)."""

tests/cron/test_bound_runner.py

@@ -0,0 +1,125 @@
+from __future__ import annotations
+
+import asyncio
+import json
+from contextlib import suppress
+from pathlib import Path
+
+import pytest
+
+from nanobot.agent.automation_turns import AutomationTurnAcceptedCancellation
+from nanobot.agent.cron_turns import CronTurnCoordinator
+from nanobot.bus.events import InboundMessage, OutboundMessage
+from nanobot.cron.bound_runner import run_bound_cron_job
+from nanobot.cron.service import CronService
+from nanobot.cron.types import CronJob, CronSchedule
+
+
+class _NoCronToolRegistry:
+    def get(self, _name: str) -> None:
+        return None
+
+
+class _CoordinatedAgent:
+    def __init__(self) -> None:
+        self.tools = _NoCronToolRegistry()
+        self.inbound: asyncio.Queue[InboundMessage] = asyncio.Queue()
+        self.coordinator = CronTurnCoordinator(
+            publish_inbound=self.inbound.put,
+            dispatch=lambda _msg: asyncio.sleep(0),
+            is_running=lambda: True,
+        )
+
+    async def submit_cron_turn(
+        self,
+        msg: InboundMessage,
+    ) -> OutboundMessage | None:
+        return await self.coordinator.submit(msg)
+
+
+@pytest.mark.asyncio
+async def test_accepted_bound_turn_cancellation_advances_durable_schedule(
+    tmp_path: Path,
+) -> None:
+    store_path = tmp_path / "cron" / "jobs.json"
+    agent = _CoordinatedAgent()
+    service = CronService(store_path)
+    job = service.add_job(
+        name="accepted reminder",
+        schedule=CronSchedule(kind="every", every_ms=60_000),
+        message="check accepted work",
+        session_key="websocket:chat-1",
+        origin_channel="websocket",
+        origin_chat_id="chat-1",
+    )
+
+    async def on_job(current: CronJob) -> str | None:
+        return await run_bound_cron_job(current, agent=agent, cron=service)
+
+    service.on_job = on_job
+    work_accepted = asyncio.Event()
+    release_work = asyncio.Event()
+    effects: list[str] = []
+
+    async def agent_worker() -> None:
+        msg = await agent.inbound.get()
+        work_accepted.set()
+        await release_work.wait()
+        effects.append(msg.content)
+        agent.coordinator.complete(msg)
+
+    worker = asyncio.create_task(agent_worker())
+    run = asyncio.create_task(service.run_job(job.id))
+    try:
+        await asyncio.wait_for(work_accepted.wait(), timeout=1)
+        run.cancel()
+        with pytest.raises(AutomationTurnAcceptedCancellation):
+            await asyncio.wait_for(run, timeout=1)
+
+        # Cancellation is prompt and the independently-owned agent turn can
+        # finish later, while the cron schedule is already durable.
+        assert effects == []
+        persisted = CronService(store_path).get_job(job.id)
+        assert persisted is not None
+        assert persisted.state.last_status == "ok"
+        assert persisted.state.last_error is None
+        assert persisted.state.last_run_at_ms is not None
+        assert persisted.state.next_run_at_ms is not None
+        assert persisted.state.next_run_at_ms > persisted.state.last_run_at_ms
+        assert len(persisted.state.run_history) == 1
+        assert persisted.state.run_history[0].status == "ok"
+
+        records = [
+            json.loads(path.read_text(encoding="utf-8"))
+            for path in (store_path.parent / "runs").glob("*.json")
+        ]
+        assert len(records) == 1
+        assert records[0]["status"] == "accepted"
+
+        replayed: list[str] = []
+
+        async def replay_job(current: CronJob) -> None:
+            replayed.append(current.id)
+
+        restarted = CronService(store_path, on_job=replay_job)
+        restarted._running = True
+        restarted._arm_timer = lambda: None
+        try:
+            await restarted._on_timer()
+        finally:
+            restarted.stop()
+        assert replayed == []
+
+        release_work.set()
+        await asyncio.wait_for(worker, timeout=1)
+        assert effects == [records[0]["rendered_prompt"]]
+    finally:
+        release_work.set()
+        if not run.done():
+            run.cancel()
+        if not worker.done():
+            worker.cancel()
+        with suppress(asyncio.CancelledError):
+            await run
+        with suppress(asyncio.CancelledError):
+            await worker

tests/cron/test_cron_service.py

@@ -1,5 +1,6 @@
 import asyncio
 import json
+import threading
 import time
 from pathlib import Path
 
@@ -653,6 +654,231 @@ async def test_manual_run_persists_completion_when_callback_lists_jobs(tmp_path)
     assert state["runHistory"][0]["status"] == "ok"
 
 
+@pytest.mark.asyncio
+async def test_same_job_manual_runs_do_not_overlap(tmp_path) -> None:
+    entered = asyncio.Event()
+    release = asyncio.Event()
+    calls: list[str] = []
+
+    async def on_job(job: CronJob) -> None:
+        calls.append(job.id)
+        entered.set()
+        await release.wait()
+
+    service = CronService(tmp_path / "cron" / "jobs.json", on_job=on_job)
+    job = service.add_job(
+        name="single-flight",
+        schedule=CronSchedule(kind="every", every_ms=60_000),
+        message="hello",
+        **_bound_chat(),
+    )
+
+    first = asyncio.create_task(service.run_job(job.id))
+    try:
+        await asyncio.wait_for(entered.wait(), timeout=1)
+        assert await service.run_job(job.id) is False
+        assert calls == [job.id]
+    finally:
+        release.set()
+        assert await first is True
+
+    loaded = service.get_job(job.id)
+    assert loaded is not None
+    assert len(loaded.state.run_history) == 1
+
+
+@pytest.mark.asyncio
+async def test_timer_and_manual_run_of_same_job_do_not_overlap(tmp_path) -> None:
+    entered = asyncio.Event()
+    release = asyncio.Event()
+    calls: list[str] = []
+
+    async def on_job(job: CronJob) -> None:
+        calls.append(job.id)
+        entered.set()
+        await release.wait()
+
+    service = CronService(tmp_path / "cron" / "jobs.json", on_job=on_job)
+    service._running = True
+    service._arm_timer = lambda: None
+    service._load_store()
+    job = service.add_job(
+        name="timer-single-flight",
+        schedule=CronSchedule(kind="every", every_ms=60_000),
+        message="hello",
+        **_bound_chat(),
+    )
+    job.state.next_run_at_ms = max(1, int(time.time() * 1000) - 1_000)
+    service._save_store()
+
+    timer = asyncio.create_task(service._on_timer())
+    try:
+        await asyncio.wait_for(entered.wait(), timeout=1)
+        assert await service.run_job(job.id, force=True) is False
+        assert calls == [job.id]
+    finally:
+        release.set()
+        await timer
+        service.stop()
+
+    loaded = service.get_job(job.id)
+    assert loaded is not None
+    assert len(loaded.state.run_history) == 1
+
+
+@pytest.mark.asyncio
+async def test_start_cancellation_waits_for_load_and_fully_starts(
+    tmp_path,
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    store_path = tmp_path / "cron" / "jobs.json"
+    service = CronService(store_path)
+    job = service.add_job(
+        name="load-settlement",
+        schedule=CronSchedule(kind="every", every_ms=60_000),
+        message="hello",
+        **_bound_chat(),
+    )
+    entered = threading.Event()
+    release = threading.Event()
+    original_load_store = service._load_store
+
+    def blocking_load_store():
+        entered.set()
+        if not release.wait(timeout=2):
+            raise TimeoutError("test did not release cron load worker")
+        return original_load_store()
+
+    monkeypatch.setattr(service, "_load_store", blocking_load_store)
+    start = asyncio.create_task(service.start())
+    assert await asyncio.to_thread(entered.wait, 1)
+
+    try:
+        assert start.cancel()
+        await asyncio.sleep(0)
+        assert not start.done()
+        assert not store_path.exists()
+
+        release.set()
+        with pytest.raises(asyncio.CancelledError):
+            await asyncio.wait_for(start, timeout=1)
+
+        assert service._running is True
+        timer = service._timer_task
+        assert timer is not None and not timer.done()
+        assert service._store_dirty is False
+        assert (tmp_path / "cron" / "action.jsonl").read_text(encoding="utf-8") == ""
+        persisted = store_path.read_bytes()
+        assert json.loads(persisted)["jobs"][0]["id"] == job.id
+
+        await asyncio.sleep(0.01)
+        assert store_path.read_bytes() == persisted
+        assert service._timer_task is timer
+    finally:
+        release.set()
+        await asyncio.gather(start, return_exceptions=True)
+        timer = service._timer_task
+        service.stop()
+        if timer is not None:
+            await asyncio.gather(timer, return_exceptions=True)
+
+
+@pytest.mark.asyncio
+async def test_start_repeated_cancellation_waits_for_persistence_and_fully_starts(
+    tmp_path,
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    store_path = tmp_path / "cron" / "jobs.json"
+    service = CronService(store_path)
+    entered = threading.Event()
+    release = threading.Event()
+    original_save_store = service._save_store
+
+    def blocking_save_store() -> None:
+        entered.set()
+        if not release.wait(timeout=2):
+            raise TimeoutError("test did not release cron persistence worker")
+        original_save_store()
+
+    monkeypatch.setattr(service, "_save_store", blocking_save_store)
+    start = asyncio.create_task(service.start())
+    assert await asyncio.to_thread(entered.wait, 1)
+
+    try:
+        assert start.cancel()
+        await asyncio.sleep(0)
+        assert not start.done()
+        assert start.cancel()
+        await asyncio.sleep(0)
+        assert not start.done()
+        assert not store_path.exists()
+
+        release.set()
+        with pytest.raises(asyncio.CancelledError):
+            await asyncio.wait_for(start, timeout=1)
+
+        assert service._running is True
+        timer = service._timer_task
+        assert timer is not None and not timer.done()
+        assert service._store_dirty is False
+        persisted = store_path.read_bytes()
+        assert json.loads(persisted) == {"version": 1, "jobs": []}
+
+        await asyncio.sleep(0.01)
+        assert store_path.read_bytes() == persisted
+        assert service._timer_task is timer
+    finally:
+        release.set()
+        await asyncio.gather(start, return_exceptions=True)
+        timer = service._timer_task
+        service.stop()
+        if timer is not None:
+            await asyncio.gather(timer, return_exceptions=True)
+
+
+@pytest.mark.asyncio
+async def test_run_sync_cancellation_drains_worker_under_transaction_lock(tmp_path) -> None:
+    service = CronService(tmp_path / "cron" / "jobs.json")
+    entered = threading.Event()
+    release = threading.Event()
+    second_entered = threading.Event()
+    mutations: list[str] = []
+
+    def blocking_mutation() -> str:
+        entered.set()
+        if not release.wait(timeout=2):
+            raise TimeoutError("test did not release cron worker")
+        mutations.append("first")
+        return "first"
+
+    def later_mutation() -> str:
+        second_entered.set()
+        mutations.append("second")
+        return "second"
+
+    first = asyncio.create_task(service.run_sync(blocking_mutation))
+    assert await asyncio.to_thread(entered.wait, 1)
+    first.cancel()
+    second = asyncio.create_task(service.run_sync(later_mutation))
+    try:
+        await asyncio.sleep(0.05)
+        assert not first.done()
+        assert not second_entered.is_set()
+
+        release.set()
+        with pytest.raises(asyncio.CancelledError):
+            await asyncio.wait_for(first, timeout=1)
+        assert mutations == ["first"]
+
+        assert await asyncio.wait_for(second, timeout=1) == "second"
+        assert mutations == ["first", "second"]
+        await asyncio.sleep(0.01)
+        assert mutations == ["first", "second"]
+    finally:
+        release.set()
+        await asyncio.gather(first, second, return_exceptions=True)
+
+
 @pytest.mark.asyncio
 async def test_overlapping_manual_runs_preserve_stopped_service_state(tmp_path) -> None:
     store_path = tmp_path / "cron" / "jobs.json"

tests/cron/test_cron_tool_schema_contract.py

@@ -9,6 +9,7 @@ and tightens the runtime error for ``add`` without ``message``.
 
 from __future__ import annotations
 
+import asyncio
 from collections.abc import Iterator
 
 import pytest
@@ -21,6 +22,9 @@ from nanobot.agent.tools.registry import ToolRegistry
 class _SvcStub:
     """Minimal CronService stand-in; we only exercise schema/dispatch paths."""
 
+    async def run_sync(self, operation, /, *args, **kwargs):
+        return await asyncio.to_thread(operation, *args, **kwargs)
+
     def list_jobs(self):
         return []
 
@@ -74,8 +78,6 @@ class TestSchemaContract:
         # Schema permits omitting message; the runtime must return a message
         # that tells the LLM exactly what's missing and how to retry, so it
         # doesn't loop like #3113 reports.
-        import asyncio
-
         tool = registry._tools["cron"]  # type: ignore[attr-defined]
         out = asyncio.run(tool.execute(action="add", at="2030-01-01T00:00:00"))
         assert "message" in out

tests/cron/test_cron_tool_worker.py

@@ -0,0 +1,82 @@
+"""Worker-thread regressions for live CronTool mutations."""
+
+import asyncio
+import json
+import threading
+
+import pytest
+
+from nanobot.agent.tools.context import RequestContext, request_context
+from nanobot.agent.tools.cron import CronTool
+from nanobot.cron.service import CronService
+
+
+@pytest.mark.asyncio
+async def test_cron_tool_add_on_started_service_is_durable_and_rearms_owner_loop(
+    tmp_path,
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    store_path = tmp_path / "cron" / "jobs.json"
+    service = CronService(store_path, max_sleep_ms=60_000)
+    await service.start()
+
+    owner_thread_id = threading.get_ident()
+    initial_timer = service._timer_task
+    worker_thread_ids: list[int] = []
+    arm_thread_ids: list[int] = []
+    timer_rearmed = asyncio.Event()
+    tool = CronTool(service)
+
+    original_execute_sync = tool._execute_sync
+
+    def tracked_execute_sync(*args, **kwargs):
+        worker_thread_ids.append(threading.get_ident())
+        return original_execute_sync(*args, **kwargs)
+
+    original_arm_timer = service._arm_timer
+
+    def tracked_arm_timer() -> None:
+        arm_thread_ids.append(threading.get_ident())
+        original_arm_timer()
+        timer_rearmed.set()
+
+    monkeypatch.setattr(tool, "_execute_sync", tracked_execute_sync)
+    monkeypatch.setattr(service, "_arm_timer", tracked_arm_timer)
+
+    try:
+        with request_context(
+            RequestContext(
+                channel="websocket",
+                chat_id="acceptance-chat",
+                session_key="websocket:acceptance-chat",
+            )
+        ):
+            result = await tool.execute(
+                action="add",
+                name="Acceptance reminder",
+                message="Check the acceptance result",
+                every_seconds=3600,
+            )
+
+        await asyncio.wait_for(timer_rearmed.wait(), timeout=1)
+
+        assert result.startswith("Created job 'Acceptance reminder'")
+        assert "no running event loop" not in result.lower()
+        assert worker_thread_ids
+        assert all(thread_id != owner_thread_id for thread_id in worker_thread_ids)
+        assert arm_thread_ids and set(arm_thread_ids) == {owner_thread_id}
+        assert service._timer_task is not None
+        assert service._timer_task is not initial_timer
+        assert not service._timer_task.done()
+
+        stored = json.loads(store_path.read_text(encoding="utf-8"))
+        assert len(stored["jobs"]) == 1
+        assert stored["jobs"][0]["name"] == "Acceptance reminder"
+        assert stored["jobs"][0]["payload"]["sessionKey"] == "websocket:acceptance-chat"
+
+        reloaded_jobs = CronService(store_path).list_jobs(include_disabled=True)
+        assert len(reloaded_jobs) == 1
+        assert reloaded_jobs[0].name == "Acceptance reminder"
+    finally:
+        service.stop()
+        await asyncio.sleep(0)

tests/security/test_security_network.py

@@ -2,13 +2,18 @@
 
 from __future__ import annotations
 
+import asyncio
 import ipaddress
 import socket
+import threading
+import time
 from unittest.mock import patch
 
 import pytest
 
 from nanobot.security.network import (
+    async_resolve_url_target,
+    async_validate_resolved_url,
     configure_ssrf_whitelist,
     contains_internal_url,
     env_proxy_applies_to_url,
@@ -362,3 +367,78 @@ def test_whitelist_allows_ipv6_mapped_cgnat():
             assert ok, f"Whitelisted IPv6-mapped CGNAT should be allowed, got: {err}"
     finally:
         configure_ssrf_whitelist([])
+
+
+@pytest.mark.asyncio
+async def test_async_dns_resolution_keeps_event_loop_responsive(
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    started = threading.Event()
+
+    def slow_resolver(hostname, port, family=0, type=0, proto=0, flags=0):  # noqa: A002
+        started.set()
+        time.sleep(0.08)
+        return [
+            (socket.AF_INET, socket.SOCK_STREAM, 0, "", ("93.184.216.34", 0)),
+        ]
+
+    monkeypatch.setattr(socket, "getaddrinfo", slow_resolver)
+    task = asyncio.create_task(async_resolve_url_target("https://example.com/path"))
+    assert await asyncio.to_thread(started.wait, 0.5)
+    for _ in range(3):
+        await asyncio.sleep(0.01)
+    assert not task.done()
+
+    assert await asyncio.wait_for(task, timeout=0.5) == (
+        True,
+        "",
+        ("93.184.216.34",),
+    )
+
+
+@pytest.mark.asyncio
+async def test_async_dns_validation_preserves_sync_security_results(
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    def resolver(hostname, port, family=0, type=0, proto=0, flags=0):  # noqa: A002
+        assert hostname == "evil.com"
+        return [
+            (socket.AF_INET, socket.SOCK_STREAM, 0, "", ("169.254.169.254", 0)),
+        ]
+
+    monkeypatch.setattr(socket, "getaddrinfo", resolver)
+
+    sync_result = resolve_url_target("http://evil.com/latest")
+    async_result = await async_resolve_url_target("http://evil.com/latest")
+    redirect_result = await async_validate_resolved_url("http://evil.com/latest")
+
+    assert async_result == sync_result
+    assert async_result[0] is False
+    assert redirect_result[0] is False
+
+
+@pytest.mark.asyncio
+async def test_async_dns_resolution_times_out(
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    loop = asyncio.get_running_loop()
+    never_resolves = asyncio.Event()
+
+    async def stalled_getaddrinfo(*args, **kwargs):
+        await never_resolves.wait()
+
+    monkeypatch.setattr(loop, "getaddrinfo", stalled_getaddrinfo)
+
+    result = await asyncio.wait_for(
+        async_resolve_url_target(
+            "https://example.com/path",
+            timeout_s=0.01,
+        ),
+        timeout=0.2,
+    )
+
+    assert result == (
+        False,
+        "Timed out resolving hostname: example.com",
+        (),
+    )

tests/session/test_recovery.py

@@ -178,13 +178,13 @@ async def test_scan_loads_only_sessions_that_need_webui_recovery(
 
     coordinator, _, restarted = _coordinator(tmp_path)
     loaded: list[str] = []
-    get_or_create = restarted.get_or_create
+    get_or_create_async = restarted.get_or_create_async
 
-    def tracked_get_or_create(key: str) -> Session:
+    async def tracked_get_or_create(key: str) -> Session:
         loaded.append(key)
-        return get_or_create(key)
+        return await get_or_create_async(key)
 
-    monkeypatch.setattr(restarted, "get_or_create", tracked_get_or_create)
+    monkeypatch.setattr(restarted, "get_or_create_async", tracked_get_or_create)
     monkeypatch.setattr(
         "nanobot.webui.transcript.has_unfinished_transcript_tail",
         lambda _key: False,

tests/session/test_session_cache.py

@@ -1,4 +1,6 @@
+import asyncio
 import gc
+import threading
 import weakref
 
 from nanobot.session.manager import SESSION_CACHE_MAX_SIZE, SessionManager
@@ -87,3 +89,39 @@ def test_transient_session_never_reaches_storage(tmp_path) -> None:
     assert list(manager.sessions_dir.glob("*.jsonl")) == []
     manager.invalidate(session.key)
     assert manager.get_cached(session.key) is None
+
+
+async def test_concurrent_first_async_gets_share_cached_identity(
+    tmp_path,
+    monkeypatch,
+) -> None:
+    manager = SessionManager(tmp_path)
+    key = "test:concurrent-first-load"
+    load_started = threading.Event()
+    release_load = threading.Event()
+    load_calls = 0
+
+    def delayed_load(loaded_key: str):
+        nonlocal load_calls
+        assert loaded_key == key
+        load_calls += 1
+        load_started.set()
+        assert release_load.wait(timeout=1)
+        return None
+
+    monkeypatch.setattr(manager, "_load", delayed_load)
+    first_task = asyncio.create_task(manager.get_or_create_async(key))
+    try:
+        assert await asyncio.to_thread(load_started.wait, 0.5)
+        second_task = asyncio.create_task(manager.get_or_create_async(key))
+        await asyncio.sleep(0)
+        assert not second_task.done()
+    finally:
+        release_load.set()
+
+    first, second = await asyncio.gather(first_task, second_task)
+
+    assert load_calls == 1
+    assert first is second
+    assert manager.get_cached(key) is first
+    assert await manager.get_or_create_async(key) is first

tests/session/test_session_cancellation.py

@@ -0,0 +1,162 @@
+"""Cancellation settlement guarantees for native SessionManager mutations."""
+
+from __future__ import annotations
+
+import asyncio
+import threading
+from pathlib import Path
+from typing import Any
+
+import pytest
+
+from nanobot.session.manager import Session, SessionManager
+
+
+async def _cancel_blocked_mutation(
+    task: asyncio.Task[Any],
+    *,
+    started: threading.Event,
+    release: threading.Event,
+) -> None:
+    assert await asyncio.to_thread(started.wait, 1)
+    try:
+        task.cancel()
+        await asyncio.sleep(0)
+        assert not task.done(), "cancellation escaped before the mutation worker settled"
+    finally:
+        release.set()
+
+    with pytest.raises(asyncio.CancelledError):
+        await asyncio.wait_for(task, timeout=1)
+
+
+async def test_save_async_cancellation_waits_for_durable_write_and_cache(
+    tmp_path: Path,
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    manager = SessionManager(tmp_path)
+    session = Session(key="test:cancel-save")
+    session.add_message("user", "persist exactly once")
+    started = threading.Event()
+    release = threading.Event()
+    finished = threading.Event()
+    save_calls = 0
+    original_save = manager._store.save
+
+    def blocked_save(target: Session, *, fsync: bool = False) -> None:
+        nonlocal save_calls
+        save_calls += 1
+        started.set()
+        assert release.wait(timeout=1)
+        original_save(target, fsync=fsync)
+        finished.set()
+
+    monkeypatch.setattr(manager._store, "save", blocked_save)
+    task = asyncio.create_task(manager.save_async(session))
+
+    await _cancel_blocked_mutation(task, started=started, release=release)
+
+    assert finished.is_set()
+    assert save_calls == 1
+    assert manager.get_cached(session.key) is session
+    durable = manager.read_session_file(session.key)
+    assert durable is not None
+    assert durable["messages"][0]["content"] == "persist exactly once"
+
+    await asyncio.sleep(0.05)
+    assert save_calls == 1
+    assert manager.read_session_file(session.key) == durable
+
+
+async def test_update_metadata_async_cancellation_waits_for_file_and_cache_refresh(
+    tmp_path: Path,
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    manager = SessionManager(tmp_path)
+    session = manager.get_or_create("test:cancel-metadata")
+    session.metadata["title"] = "before"
+    manager.save(session)
+    started = threading.Event()
+    release = threading.Event()
+    finished = threading.Event()
+    update_calls = 0
+    original_update = manager._store.update_metadata
+
+    def blocked_update(
+        key: str,
+        updates: dict[str, Any],
+        *,
+        fsync: bool = False,
+    ) -> bool:
+        nonlocal update_calls
+        update_calls += 1
+        started.set()
+        assert release.wait(timeout=1)
+        updated = original_update(key, updates, fsync=fsync)
+        finished.set()
+        return updated
+
+    monkeypatch.setattr(manager._store, "update_metadata", blocked_update)
+    task = asyncio.create_task(
+        manager.update_session_metadata_async(
+            session.key,
+            {"title": "after", "settled": True},
+        )
+    )
+
+    await _cancel_blocked_mutation(task, started=started, release=release)
+
+    assert finished.is_set()
+    assert update_calls == 1
+    assert session.metadata["title"] == "after"
+    assert session.metadata["settled"] is True
+    durable = manager.read_session_file(session.key)
+    assert durable is not None
+    assert durable["metadata"]["title"] == "after"
+    assert durable["metadata"]["settled"] is True
+
+    await asyncio.sleep(0.05)
+    assert update_calls == 1
+    assert manager.read_session_file(session.key) == durable
+
+
+async def test_delete_async_cancellation_waits_for_file_cache_and_observer(
+    tmp_path: Path,
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    manager = SessionManager(tmp_path)
+    session = manager.get_or_create("test:cancel-delete")
+    session.add_message("user", "delete me")
+    manager.save(session)
+    observed: list[str] = []
+    manager.set_delete_observer(observed.append)
+    started = threading.Event()
+    release = threading.Event()
+    finished = threading.Event()
+    delete_calls = 0
+    original_delete = manager._store.delete
+
+    def blocked_delete(key: str) -> bool:
+        nonlocal delete_calls
+        delete_calls += 1
+        started.set()
+        assert release.wait(timeout=1)
+        deleted = original_delete(key)
+        finished.set()
+        return deleted
+
+    monkeypatch.setattr(manager._store, "delete", blocked_delete)
+    task = asyncio.create_task(manager.delete_session_async(session.key))
+
+    await _cancel_blocked_mutation(task, started=started, release=release)
+
+    assert finished.is_set()
+    assert delete_calls == 1
+    assert manager.read_session_file(session.key) is None
+    assert manager.get_cached(session.key) is None
+    assert observed == [session.key]
+
+    await asyncio.sleep(0.05)
+    assert delete_calls == 1
+    assert observed == [session.key]
+    assert manager.read_session_file(session.key) is None

tests/test_event_loop_responsiveness.py

@@ -0,0 +1,127 @@
+"""Cross-cutting heartbeat regressions for synchronous persistence locks."""
+
+from __future__ import annotations
+
+import asyncio
+from contextlib import suppress
+from pathlib import Path
+
+from filelock import FileLock
+
+from nanobot.agent.tools.context import RequestContext, request_context
+from nanobot.agent.tools.cron import CronTool
+from nanobot.config.loader import load_config, save_config
+from nanobot.config.schema import Config
+from nanobot.cron.service import CronService
+from nanobot.session.manager import SessionManager
+from nanobot.triggers.local_runner import run_local_trigger_queue
+from nanobot.triggers.local_store import LocalTriggerStore
+from nanobot.webui.settings_api import update_api_settings
+from nanobot.webui.settings_services import WebUISettingsServices
+
+
+async def _assert_10ms_heartbeat_runs_while_pending(task: asyncio.Task[object]) -> None:
+    for _ in range(3):
+        await asyncio.sleep(0.01)
+    assert not task.done()
+
+
+async def test_session_lock_contention_does_not_block_event_loop(
+    tmp_path: Path,
+    monkeypatch,
+) -> None:
+    workspace = tmp_path / "workspace"
+    sessions_root = tmp_path / "session-data"
+    legacy_root = tmp_path / "legacy-sessions"
+    monkeypatch.setattr(
+        "nanobot.session.manager.get_legacy_sessions_dir",
+        lambda: legacy_root,
+    )
+    manager = SessionManager(workspace, sessions_root=sessions_root)
+    session = manager.get_or_create("websocket:lock-test")
+    session.add_message("user", "hello")
+    lock = manager._jsonl_store._session_files_lock
+    assert lock.timeout == 5
+    blocker = FileLock(lock.lock_file, timeout=1)
+    blocker.acquire()
+    task = asyncio.create_task(manager.save_async(session))
+    try:
+        await _assert_10ms_heartbeat_runs_while_pending(task)
+    finally:
+        blocker.release()
+
+    await asyncio.wait_for(task, timeout=1)
+    assert manager.read_session_file(session.key) is not None
+
+
+async def test_cron_lock_contention_does_not_block_event_loop(tmp_path: Path) -> None:
+    service = CronService(tmp_path / "cron" / "jobs.json")
+    tool = CronTool(service)
+    assert service._lock.timeout == 5
+    blocker = FileLock(service._lock.lock_file, timeout=1)
+    blocker.acquire()
+    with request_context(
+        RequestContext(
+            channel="websocket",
+            chat_id="lock-test",
+            session_key="websocket:lock-test",
+        )
+    ):
+        task = asyncio.create_task(
+            tool.execute(action="add", message="wake up", every_seconds=60)
+        )
+    try:
+        await _assert_10ms_heartbeat_runs_while_pending(task)
+    finally:
+        blocker.release()
+
+    result = await asyncio.wait_for(task, timeout=1)
+    assert result.startswith("Created job")
+
+
+async def test_local_trigger_lock_contention_does_not_block_event_loop(
+    tmp_path: Path,
+) -> None:
+    store = LocalTriggerStore(tmp_path)
+    assert store._lock.timeout == 5
+    blocker = FileLock(store._lock.lock_file, timeout=1)
+    blocker.acquire()
+
+    async def submit_turn(_message):
+        return None
+
+    task = asyncio.create_task(
+        run_local_trigger_queue(
+            store=store,
+            submit_turn=submit_turn,
+            is_channel_enabled=lambda _name: True,
+            poll_interval_s=0.01,
+        )
+    )
+    try:
+        await _assert_10ms_heartbeat_runs_while_pending(task)
+    finally:
+        blocker.release()
+    await asyncio.sleep(0.05)
+    task.cancel()
+    with suppress(asyncio.CancelledError):
+        await asyncio.wait_for(task, timeout=1)
+
+
+async def test_settings_lock_contention_does_not_block_event_loop(tmp_path: Path) -> None:
+    config_path = tmp_path / "config.json"
+    save_config(Config(), config_path)
+    services = WebUISettingsServices.create(config_path)
+    assert services.config._file_lock.timeout == 5
+    blocker = FileLock(services.config._file_lock.lock_file, timeout=1)
+    blocker.acquire()
+    task = asyncio.create_task(
+        services.mutate_async(update_api_settings, {"port": ["19001"]})
+    )
+    try:
+        await _assert_10ms_heartbeat_runs_while_pending(task)
+    finally:
+        blocker.release()
+
+    await asyncio.wait_for(task, timeout=1)
+    assert load_config(config_path).api.port == 19001

tests/tools/test_exec_security.py

@@ -2,8 +2,10 @@
 
 from __future__ import annotations
 
+import asyncio
 import socket
 import sys
+import threading
 from unittest.mock import patch
 
 import pytest
@@ -173,6 +175,37 @@ async def test_exec_blocks_chained_internal_url():
     assert "Error" in result
 
 
+@pytest.mark.asyncio
+async def test_restricted_exec_url_guard_keeps_event_loop_responsive(
+    tmp_path,
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    resolver_started = threading.Event()
+    release_resolver = threading.Event()
+
+    def slow_private_resolver(hostname, port, family=0, type=0, proto=0, flags=0):  # noqa: A002
+        resolver_started.set()
+        assert release_resolver.wait(timeout=1)
+        return _fake_resolve_private(hostname, port, family, type)
+
+    monkeypatch.setattr(socket, "getaddrinfo", slow_private_resolver)
+    tool = ExecTool(working_dir=str(tmp_path), restrict_to_workspace=True)
+    task = asyncio.create_task(tool.execute(command="curl https://example.com/latest"))
+    safety_release = threading.Timer(0.2, release_resolver.set)
+    safety_release.start()
+    try:
+        assert await asyncio.to_thread(resolver_started.wait, 0.5)
+        for _ in range(3):
+            await asyncio.sleep(0)
+        assert not task.done()
+    finally:
+        release_resolver.set()
+        safety_release.cancel()
+
+    result = await asyncio.wait_for(task, timeout=0.5)
+    assert "internal/private URL detected" in result
+
+
 # --- #2989: block writes to nanobot internal state files -----------------
 
 

tests/tools/test_filesystem_tools.py

@@ -1,7 +1,12 @@
 """Tests for enhanced filesystem tools: ReadFileTool, EditFileTool, ListDirTool."""
 
+import asyncio
+import threading
+
 import pytest
 
+from nanobot.agent.tools import file_state
+from nanobot.agent.tools import filesystem as filesystem_tools
 from nanobot.agent.tools.filesystem import (
     EditFileTool,
     ListDirTool,
@@ -478,3 +483,188 @@ class TestWorkspaceRestriction:
         )
         assert "Successfully edited" in result
         assert target.read_text(encoding="utf-8") == "after\n"
+
+
+@pytest.mark.asyncio
+async def test_edit_matching_keeps_event_loop_responsive(tmp_path, monkeypatch):
+    target = tmp_path / "slow.py"
+    target.write_text("before\n", encoding="utf-8")
+    tool = EditFileTool(workspace=tmp_path)
+    original_find_matches = filesystem_tools._find_matches
+    started = threading.Event()
+    release = threading.Event()
+
+    def blocking_find_matches(*args, **kwargs):
+        started.set()
+        if not release.wait(timeout=1):
+            raise TimeoutError("test did not release edit matching")
+        return original_find_matches(*args, **kwargs)
+
+    monkeypatch.setattr(filesystem_tools, "_find_matches", blocking_find_matches)
+    task = asyncio.create_task(
+        tool.execute(path=str(target), old_text="before", new_text="after")
+    )
+    try:
+        assert await asyncio.to_thread(started.wait, 0.5)
+        for _ in range(3):
+            await asyncio.sleep(0.01)
+        assert not task.done()
+    finally:
+        release.set()
+
+    assert "Successfully edited" in await asyncio.wait_for(task, timeout=0.5)
+    assert target.read_text(encoding="utf-8") == "after\n"
+
+
+@pytest.mark.asyncio
+async def test_edit_cancellation_before_commit_prevents_delayed_write(tmp_path, monkeypatch):
+    target = tmp_path / "cancel.py"
+    target.write_text("before\n", encoding="utf-8")
+    tool = EditFileTool(workspace=tmp_path)
+    original_find_matches = filesystem_tools._find_matches
+    started = threading.Event()
+    release = threading.Event()
+
+    def blocking_find_matches(*args, **kwargs):
+        started.set()
+        if not release.wait(timeout=1):
+            raise TimeoutError("test did not release edit matching")
+        return original_find_matches(*args, **kwargs)
+
+    monkeypatch.setattr(filesystem_tools, "_find_matches", blocking_find_matches)
+    task = asyncio.create_task(
+        tool.execute(path=str(target), old_text="before", new_text="after")
+    )
+    assert await asyncio.to_thread(started.wait, 0.5)
+    task.cancel()
+    with pytest.raises(asyncio.CancelledError):
+        await asyncio.wait_for(task, timeout=0.2)
+    release.set()
+    await asyncio.sleep(0.05)
+
+    assert target.read_text(encoding="utf-8") == "before\n"
+
+
+@pytest.mark.asyncio
+async def test_edit_repeated_cancellation_during_commit_waits_for_settlement(
+    tmp_path,
+    monkeypatch,
+):
+    target = tmp_path / "commit.py"
+    target.write_text("before\n", encoding="utf-8")
+    tool = EditFileTool(workspace=tmp_path)
+    path_type = type(target)
+    original_write_bytes = path_type.write_bytes
+    original_wait_for_commit = EditFileTool._wait_for_commit
+    commit_started = threading.Event()
+    drain_started = threading.Event()
+    release = threading.Event()
+    writes: list[bytes] = []
+
+    def blocking_write_bytes(path, data):
+        if path == target:
+            commit_started.set()
+            if not release.wait(timeout=1):
+                raise TimeoutError("test did not release edit commit")
+            writes.append(data)
+        return original_write_bytes(path, data)
+
+    def observed_wait_for_commit(commit_lock):
+        drain_started.set()
+        original_wait_for_commit(commit_lock)
+
+    monkeypatch.setattr(path_type, "write_bytes", blocking_write_bytes)
+    monkeypatch.setattr(
+        EditFileTool,
+        "_wait_for_commit",
+        staticmethod(observed_wait_for_commit),
+    )
+    task = asyncio.create_task(
+        tool.execute(path=str(target), old_text="before", new_text="after")
+    )
+    try:
+        assert await asyncio.to_thread(commit_started.wait, 0.5)
+        assert task.cancel()
+        assert await asyncio.to_thread(drain_started.wait, 0.5)
+
+        # Later cancellation requests must not interrupt the in-flight commit drain.
+        assert task.cancel()
+        await asyncio.sleep(0)
+        assert not task.done()
+        assert task.cancel()
+        await asyncio.sleep(0)
+        assert not task.done()
+        assert target.read_text(encoding="utf-8") == "before\n"
+    finally:
+        release.set()
+
+    with pytest.raises(asyncio.CancelledError):
+        await asyncio.wait_for(task, timeout=0.5)
+
+    final_content = target.read_bytes()
+    final_state = tool._file_states.get(target)
+    assert target.read_text(encoding="utf-8") == "after\n"
+    assert writes == [final_content]
+    assert final_state is not None
+    assert tool._file_states.check_read(target) is None
+
+    await asyncio.sleep(0.05)
+    assert target.read_bytes() == final_content
+    assert tool._file_states.get(target) is final_state
+    assert writes == [final_content]
+
+
+def test_exact_match_line_accounting_is_linear_and_occurrence_can_stop_early():
+    class CountingText(str):
+        count_span = 0
+        find_calls = 0
+
+        def count(self, sub, start=None, end=None):
+            actual_start = 0 if start is None else start
+            actual_end = len(self) if end is None else end
+            self.count_span += max(0, actual_end - actual_start)
+            return super().count(sub, actual_start, actual_end)
+
+        def find(self, sub, start=None, end=None):
+            self.find_calls += 1
+            actual_start = 0 if start is None else start
+            actual_end = len(self) if end is None else end
+            return super().find(sub, actual_start, actual_end)
+
+    content = CountingText("match\n" * 10_000)
+    matches = filesystem_tools._find_exact_matches(content, "match")
+
+    assert len(matches) == 10_000
+    assert matches[-1].line == 10_000
+    assert content.count_span <= len(content)
+
+    first_only = CountingText(content)
+    matches = filesystem_tools._find_matches(
+        first_only,
+        "match",
+        max_exact_matches=1,
+    )
+    assert len(matches) == 1
+    assert first_only.find_calls == 1
+
+
+@pytest.mark.asyncio
+async def test_edit_worker_preserves_bound_file_state_context(tmp_path):
+    target = tmp_path / "context.py"
+    target.write_text("before\n", encoding="utf-8")
+    states = file_state.FileStates()
+    states.record_read(target)
+    tool = EditFileTool(workspace=tmp_path)
+    token = file_state.bind_file_states(states)
+    try:
+        result = await tool.execute(
+            path=str(target),
+            old_text="before",
+            new_text="after",
+        )
+    finally:
+        file_state.reset_file_states(token)
+
+    assert result == f"Successfully edited {target}"
+    assert states.get(target) is not None
+    assert target.read_text(encoding="utf-8") == "after\n"

tests/tools/test_mcp_probe.py

@@ -2,7 +2,6 @@
 from __future__ import annotations
 
 import asyncio
-import socket
 from unittest.mock import MagicMock, patch
 
 import httpx
@@ -56,10 +55,12 @@ async def test_probe_uses_default_port_for_http(monkeypatch: pytest.MonkeyPatch)
     """When no port is present, probe the validated address on port 80."""
     attempts: list[tuple[str, int]] = []
 
-    monkeypatch.setattr(
+    async def _resolve_url_target(
-        "nanobot.agent.tools.mcp.resolve_url_target",
+        _url: str,
-        lambda _url: (True, "", ("93.184.216.34",)),
+    ) -> tuple[bool, str, tuple[str, ...]]:
-    )
+        return True, "", ("93.184.216.34",)
+
+    monkeypatch.setattr(mcp_mod, "async_resolve_url_target", _resolve_url_target)
 
     async def _open_connection(host: str, port: int):
         attempts.append((host, port))
@@ -72,32 +73,43 @@ async def test_probe_uses_default_port_for_http(monkeypatch: pytest.MonkeyPatch)
 
 
 @pytest.mark.asyncio
-async def test_probe_rejects_public_name_resolving_to_loopback():
+async def test_probe_rejects_public_name_resolving_to_loopback(
-    def _resolver(hostname, port, family=0, type_=0):
+    monkeypatch: pytest.MonkeyPatch,
-        return [(socket.AF_INET, socket.SOCK_STREAM, 0, "", ("127.0.0.1", 0))]
+) -> None:
+    async def _resolve_url_target(
+        _url: str,
+    ) -> tuple[bool, str, tuple[str, ...]]:
+        return False, "Blocked: example.com resolves to private/internal address 127.0.0.1", ()
+
+    monkeypatch.setattr(mcp_mod, "async_resolve_url_target", _resolve_url_target)
 
-    with patch("nanobot.security.network.socket.getaddrinfo", _resolver):
     assert await _probe_http_url("http://example.com:8765/mcp") is False
 
 
 @pytest.mark.asyncio
-async def test_probe_skips_direct_tcp_when_global_proxy_env_is_set(monkeypatch):
+async def test_probe_skips_direct_tcp_when_global_proxy_env_is_set(
-    def _resolver(hostname, port, family=0, type_=0):
+    monkeypatch: pytest.MonkeyPatch,
-        return [(socket.AF_INET, socket.SOCK_STREAM, 0, "", ("93.184.216.34", 0))]
+) -> None:
+    async def _resolve_url_target(
+        _url: str,
+    ) -> tuple[bool, str, tuple[str, ...]]:
+        return True, "", ("93.184.216.34",)
 
     async def _open_connection(*args, **kwargs):
         raise AssertionError("global proxy env should skip direct TCP probe")
 
     monkeypatch.setenv("HTTPS_PROXY", "http://proxy.example:8080")
     monkeypatch.setenv("NO_PROXY", "localhost,127.0.0.1,::1")
+    monkeypatch.setattr(mcp_mod, "async_resolve_url_target", _resolve_url_target)
     monkeypatch.setattr("nanobot.agent.tools.mcp.asyncio.open_connection", _open_connection)
 
-    with patch("nanobot.security.network.socket.getaddrinfo", _resolver):
     assert await _probe_http_url("https://mcp.example.com/mcp") is True
 
 
 @pytest.mark.asyncio
-async def test_probe_tries_next_validated_ip_when_first_is_unreachable(monkeypatch):
+async def test_probe_tries_next_validated_ip_when_first_is_unreachable(
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
     attempts: list[tuple[str, int]] = []
 
     class FakeWriter:
@@ -107,11 +119,10 @@ async def test_probe_tries_next_validated_ip_when_first_is_unreachable(monkeypat
         async def wait_closed(self):
             return None
 
-    def _resolver(hostname, port, family=0, type_=0):
+    async def _resolve_url_target(
-        return [
+        _url: str,
-            (socket.AF_INET, socket.SOCK_STREAM, 0, "", ("93.184.216.34", 0)),
+    ) -> tuple[bool, str, tuple[str, ...]]:
-            (socket.AF_INET, socket.SOCK_STREAM, 0, "", ("93.184.216.35", 0)),
+        return True, "", ("93.184.216.34", "93.184.216.35")
-        ]
 
     async def _open_connection(host: str, port: int):
         attempts.append((host, port))
@@ -119,7 +130,7 @@ async def test_probe_tries_next_validated_ip_when_first_is_unreachable(monkeypat
             raise OSError("first address unreachable")
         return object(), FakeWriter()
 
-    monkeypatch.setattr("nanobot.security.network.socket.getaddrinfo", _resolver)
+    monkeypatch.setattr(mcp_mod, "async_resolve_url_target", _resolve_url_target)
     monkeypatch.setattr("nanobot.agent.tools.mcp.asyncio.open_connection", _open_connection)
 
     assert await _probe_http_url("http://mcp.example:8765/mcp") is True
@@ -182,10 +193,13 @@ async def test_connect_isolates_streamable_http_status_failure(
     async def _reachable(_url: str) -> bool:
         return True
 
+    async def _validate_url_target(_url: str) -> tuple[bool, str]:
+        return True, ""
+
     def _return_http_530(request: httpx.Request) -> httpx.Response:
         return httpx.Response(530, text="cloudflare error 1033", request=request)
 
-    monkeypatch.setattr(mcp_mod, "validate_url_target", lambda _url: (True, ""))
+    monkeypatch.setattr(mcp_mod, "async_validate_url_target", _validate_url_target)
     monkeypatch.setattr(mcp_mod, "_probe_http_url", _reachable)
     monkeypatch.setattr(
         mcp_mod,

tests/tools/test_mcp_tool.py

@@ -213,7 +213,7 @@ async def test_saved_oauth_http_403_projects_failed_runtime_without_details(
     oauth_mod.create_mcp_oauth_auth = create_auth  # type: ignore[attr-defined]
     oauth_mod.mcp_oauth_has_credentials = lambda _name, _url: True  # type: ignore[attr-defined]
     monkeypatch.setitem(sys.modules, "nanobot.agent.tools.mcp_oauth", oauth_mod)
-    monkeypatch.setattr(mcp_mod, "validate_url_target", lambda _url: (True, ""))
+    monkeypatch.setattr(mcp_mod, "async_validate_url_target", _allow_url)
     monkeypatch.setattr(mcp_mod, "_probe_http_url", reachable)
     monkeypatch.setattr(
         sys.modules["mcp.client.streamable_http"],
@@ -1006,7 +1006,7 @@ async def test_connect_mcp_servers_env_proxy_adds_proxy_mounts_and_keeps_pinned_
     async def _reachable(_url: str) -> bool:
         return True
 
-    def _validate(_url: str) -> tuple[bool, str]:
+    async def _validate(_url: str) -> tuple[bool, str]:
         return True, ""
 
     class FakeAsyncClient:
@@ -1033,7 +1033,7 @@ async def test_connect_mcp_servers_env_proxy_adds_proxy_mounts_and_keeps_pinned_
 
     monkeypatch.setenv("HTTPS_PROXY", "http://proxy.example:8080")
     monkeypatch.setenv("NO_PROXY", "localhost,127.0.0.1,::1")
-    monkeypatch.setattr(mcp_mod, "validate_url_target", _validate)
+    monkeypatch.setattr(mcp_mod, "async_validate_url_target", _validate)
     monkeypatch.setattr(mcp_mod, "_probe_http_url", _reachable)
     monkeypatch.setattr(
         mcp_mod,
@@ -1105,7 +1105,7 @@ async def test_connect_mcp_servers_http_clients_reject_unsafe_redirect_targets(
     sent_urls: list[str] = []
     used_transports: list[str] = []
 
-    def _validate(url: str, **_kwargs: object) -> tuple[bool, str]:
+    async def _validate(url: str, **_kwargs: object) -> tuple[bool, str]:
         checked_urls.append(url)
         if url == "http://127.0.0.1/private":
             return False, "loopback blocked"
@@ -1145,7 +1145,7 @@ async def test_connect_mcp_servers_http_clients_reject_unsafe_redirect_targets(
         await http_client.get("https://example.com/start")
         yield object(), object(), object()
 
-    monkeypatch.setattr(mcp_mod, "validate_url_target", _validate)
+    monkeypatch.setattr(mcp_mod, "async_validate_url_target", _validate)
     monkeypatch.setattr(mcp_mod, "_probe_http_url", _reachable)
     monkeypatch.setattr(
         mcp_mod,
@@ -1320,7 +1320,7 @@ async def test_connect_mcp_servers_streamable_http_uses_finite_timeout(
     async def _reachable(_url: str) -> bool:
         return True
 
-    def _validate(_url: str) -> tuple[bool, str]:
+    async def _validate(_url: str) -> tuple[bool, str]:
         return True, ""
 
     @asynccontextmanager
@@ -1328,7 +1328,7 @@ async def test_connect_mcp_servers_streamable_http_uses_finite_timeout(
         captured["timeout"] = http_client.timeout
         yield object(), object(), object()
 
-    monkeypatch.setattr(mcp_mod, "validate_url_target", _validate)
+    monkeypatch.setattr(mcp_mod, "async_validate_url_target", _validate)
     monkeypatch.setattr(mcp_mod, "_probe_http_url", _reachable)
     monkeypatch.setattr(
         mcp_mod,
@@ -1371,7 +1371,7 @@ async def test_connect_mcp_servers_attaches_oauth_to_remote_http_client(
     async def _reachable(_url: str) -> bool:
         return True
 
-    def _validate(_url: str) -> tuple[bool, str]:
+    async def _validate(_url: str) -> tuple[bool, str]:
         return True, ""
 
     async def _create_auth(name: str, url: str, handlers: object) -> object:
@@ -1407,7 +1407,7 @@ async def test_connect_mcp_servers_attaches_oauth_to_remote_http_client(
         assert http_client is not None
         yield object(), object(), object()
 
-    monkeypatch.setattr(mcp_mod, "validate_url_target", _validate)
+    monkeypatch.setattr(mcp_mod, "async_validate_url_target", _validate)
     monkeypatch.setattr(mcp_mod, "_probe_http_url", _reachable)
     monkeypatch.setattr(mcp_mod.httpx, "AsyncClient", FakeAsyncClient)
     monkeypatch.setattr(sys.modules["mcp.client.sse"], "sse_client", _capturing_sse_client)
@@ -1461,7 +1461,7 @@ async def test_connect_mcp_servers_skips_background_oauth_without_credentials(
     oauth_mod.MCPAuthorizationRequiredError = AuthorizationRequiredError  # type: ignore[attr-defined]
     oauth_mod.create_mcp_oauth_auth = _create_auth  # type: ignore[attr-defined]
     monkeypatch.setitem(sys.modules, "nanobot.agent.tools.mcp_oauth", oauth_mod)
-    monkeypatch.setattr(mcp_mod, "validate_url_target", lambda _url: (True, ""))
+    monkeypatch.setattr(mcp_mod, "async_validate_url_target", _allow_url)
     monkeypatch.setattr(mcp_mod, "_probe_http_url", _probe)
 
     stacks = await connect_mcp_servers(
@@ -1959,3 +1959,12 @@ def test_long_server_name_tools_are_matched_by_server_name() -> None:
     assert removed == 1
     assert wrapper.name not in registry.tool_names
     assert other_wrapper.name in registry.tool_names
+
+
+async def _allow_url(
+    _url: str,
+    *,
+    allow_loopback: bool = False,
+) -> tuple[bool, str]:
+    del allow_loopback
+    return True, ""

tests/tools/test_search_tools.py

@@ -2,7 +2,9 @@
 
 from __future__ import annotations
 
+import asyncio
 import os
+import threading
 import time
 from pathlib import Path
 from types import SimpleNamespace
@@ -97,6 +99,82 @@ async def test_find_files_rejects_paths_outside_workspace(tmp_path: Path) -> Non
     assert result.startswith("Error:")
 
 
+@pytest.mark.asyncio
+async def test_find_files_scan_keeps_event_loop_responsive(
+    tmp_path: Path,
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    target = tmp_path / "match.txt"
+    target.write_text("ok\n", encoding="utf-8")
+    tool = FindFilesTool(workspace=tmp_path, allowed_dir=tmp_path)
+    original_iter_paths = tool._iter_paths
+    started = threading.Event()
+    release = threading.Event()
+
+    def blocking_iter_paths(root: Path, *, include_dirs: bool):
+        started.set()
+        if not release.wait(timeout=1):
+            raise TimeoutError("test did not release find_files traversal")
+        yield from original_iter_paths(root, include_dirs=include_dirs)
+
+    monkeypatch.setattr(tool, "_iter_paths", blocking_iter_paths)
+    task = asyncio.create_task(tool.execute(path="."))
+    try:
+        assert await asyncio.to_thread(started.wait, 0.5)
+        for _ in range(3):
+            await asyncio.sleep(0.01)
+        assert not task.done()
+    finally:
+        release.set()
+
+    assert await asyncio.wait_for(task, timeout=0.5) == "match.txt"
+
+
+@pytest.mark.asyncio
+async def test_find_files_cancellation_is_prompt_and_stops_scan_before_next_path(
+    tmp_path: Path,
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    (tmp_path / "match.txt").write_text("ok\n", encoding="utf-8")
+    tool = FindFilesTool(workspace=tmp_path, allowed_dir=tmp_path)
+    original_iter_paths = tool._iter_paths
+    started = threading.Event()
+    release = threading.Event()
+
+    def blocking_iter_paths(root: Path, *, include_dirs: bool):
+        started.set()
+        if not release.wait(timeout=1):
+            raise TimeoutError("test did not release find_files traversal")
+        yield from original_iter_paths(root, include_dirs=include_dirs)
+
+    monkeypatch.setattr(tool, "_iter_paths", blocking_iter_paths)
+    task = asyncio.create_task(tool.execute(path="."))
+    assert await asyncio.to_thread(started.wait, 0.5)
+    task.cancel()
+    try:
+        with pytest.raises(asyncio.CancelledError):
+            await asyncio.wait_for(task, timeout=0.2)
+        await asyncio.sleep(0.01)
+    finally:
+        release.set()
+        await asyncio.sleep(0.05)
+
+
+@pytest.mark.asyncio
+async def test_find_files_enforces_path_budget(
+    tmp_path: Path,
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    (tmp_path / "a.txt").write_text("a\n", encoding="utf-8")
+    (tmp_path / "b.txt").write_text("b\n", encoding="utf-8")
+    monkeypatch.setattr(FindFilesTool, "_MAX_SCAN_PATHS", 1)
+    tool = FindFilesTool(workspace=tmp_path, allowed_dir=tmp_path)
+
+    result = await tool.execute(path=".")
+
+    assert result.startswith("Error: find_files scan exceeded 1 paths")
+
+
 @pytest.mark.asyncio
 async def test_grep_respects_glob_filter_and_context(tmp_path: Path) -> None:
     (tmp_path / "src").mkdir()

tests/tools/test_web_fetch_jina_privacy.py

@@ -16,7 +16,7 @@ from nanobot.agent.tools.web import (
 )
 
 
-def _fake_resolve_public(hostname, port, family=0, type_=0):
+def _fake_resolve_public(hostname, port, family=0, type_=0, proto=0, flags=0):
     return [(socket.AF_INET, socket.SOCK_STREAM, 0, "", ("93.184.216.34", 0))]
 
 

tests/tools/test_web_fetch_security.py

@@ -30,11 +30,11 @@ def _clear_proxy_env(monkeypatch: pytest.MonkeyPatch) -> None:
         monkeypatch.delenv(name, raising=False)
 
 
-def _fake_resolve_private(hostname, port, family=0, type_=0):
+def _fake_resolve_private(hostname, port, family=0, type_=0, proto=0, flags=0):
     return [(socket.AF_INET, socket.SOCK_STREAM, 0, "", ("169.254.169.254", 0))]
 
 
-def _fake_resolve_public(hostname, port, family=0, type_=0):
+def _fake_resolve_public(hostname, port, family=0, type_=0, proto=0, flags=0):
     return [(socket.AF_INET, socket.SOCK_STREAM, 0, "", ("93.184.216.34", 0))]
 
 
@@ -99,7 +99,7 @@ async def test_web_fetch_blocks_private_ip():
 @pytest.mark.asyncio
 async def test_web_fetch_blocks_localhost():
     tool = WebFetchTool()
-    def _resolve_localhost(hostname, port, family=0, type_=0):
+    def _resolve_localhost(hostname, port, family=0, type_=0, proto=0, flags=0):
         return [(socket.AF_INET, socket.SOCK_STREAM, 0, "", ("127.0.0.1", 0))]
     with patch("nanobot.security.network.socket.getaddrinfo", _resolve_localhost):
         result = await tool.execute(url="http://localhost/admin")
@@ -112,7 +112,7 @@ async def test_web_fetch_blocks_localhost_even_in_full_workspace_scope(tmp_path)
     tool = WebFetchTool()
     scope = build_workspace_scope(tmp_path, "full")
 
-    def _resolve_localhost(hostname, port, family=0, type_=0):
+    def _resolve_localhost(hostname, port, family=0, type_=0, proto=0, flags=0):
         return [(socket.AF_INET, socket.SOCK_STREAM, 0, "", ("127.0.0.1", 0))]
 
     token = bind_workspace_scope(scope)
@@ -439,7 +439,7 @@ async def test_web_fetch_blocks_private_redirect_before_readability_request(monk
     monkeypatch.setattr(web_module.httpx, "AsyncClient", FakeClient)
     monkeypatch.setattr(web_module, "_pinned_dns_transport", lambda: object())
 
-    def resolve_public_start_only(hostname, port, family=0, type_=0):
+    def resolve_public_start_only(hostname, port, family=0, type_=0, proto=0, flags=0):
         if hostname == "attacker.example":
             return _fake_resolve_public(hostname, port, family, type_)
         return _REAL_GETADDRINFO(hostname, port, family, type_)
@@ -485,7 +485,7 @@ async def test_web_fetch_blocks_private_redirect_before_returning_image(monkeypa
     monkeypatch.setattr("nanobot.agent.tools.web.httpx.AsyncClient", TransportAsyncClient)
     monkeypatch.setattr(web_module, "_pinned_dns_transport", lambda: object())
 
-    def resolve_public_start_only(hostname, port, family=0, type_=0):
+    def resolve_public_start_only(hostname, port, family=0, type_=0, proto=0, flags=0):
         if hostname == "example.com":
             return _fake_resolve_public(hostname, port, family, type_)
         return _REAL_GETADDRINFO(hostname, port, family, type_)
@@ -526,7 +526,7 @@ async def test_web_fetch_does_not_request_private_redirect_target(monkeypatch):
     monkeypatch.setattr(web_module.httpx, "AsyncClient", TransportAsyncClient)
     monkeypatch.setattr(web_module, "_pinned_dns_transport", lambda: object())
 
-    def resolve_public_start_only(hostname, port, family=0, type_=0):
+    def resolve_public_start_only(hostname, port, family=0, type_=0, proto=0, flags=0):
         if hostname == "attacker.example":
             return _fake_resolve_public(hostname, port, family, type_)
         return _REAL_GETADDRINFO(hostname, port, family, type_)

tests/tools/test_web_fetch_url_sanitization.py

@@ -11,7 +11,7 @@ import pytest
 from nanobot.agent.tools.web import WebFetchTool, _validate_url
 
 
-def _fake_resolve_public(hostname, port, family=0, type_=0):
+def _fake_resolve_public(hostname, port, family=0, type_=0, proto=0, flags=0):
     import socket
     return [(socket.AF_INET, socket.SOCK_STREAM, 0, "", ("93.184.216.34", 0))]
 

tests/triggers/test_local_triggers.py

@@ -4,6 +4,7 @@ import asyncio
 import errno
 import json
 import os
+import threading
 from contextlib import suppress
 from pathlib import Path
 
@@ -13,6 +14,7 @@ from nanobot.agent.automation_turns import AutomationTurnError
 from nanobot.bus.events import InboundMessage, OutboundMessage
 from nanobot.triggers.local_runner import run_local_trigger_queue
 from nanobot.triggers.local_store import LocalTriggerStore, TriggerDisabledError
+from nanobot.triggers.local_turns import LocalTriggerTurnCoordinator
 from nanobot.triggers.local_types import LocalTrigger, TriggerDelivery
 from nanobot.webui.metadata import WEBUI_MESSAGE_SOURCE_METADATA_KEY, WEBUI_TURN_METADATA_KEY
 
@@ -276,6 +278,163 @@ def test_recover_processing_deliveries_requeues_claimed_delivery(tmp_path: Path)
     assert reclaimed[0].last_error == "delivery was recovered from interrupted processing"
 
 
+@pytest.mark.asyncio
+async def test_local_trigger_queue_cancellation_waits_for_claim_mutation(
+    tmp_path: Path,
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    store = LocalTriggerStore(tmp_path)
+    trigger = store.create(
+        name="PR review",
+        channel="websocket",
+        chat_id="chat-1",
+        session_key="websocket:chat-1",
+    )
+    store.enqueue(trigger.id, "Review PR #4591")
+    claim_started = threading.Event()
+    allow_claim = threading.Event()
+    claim_finished = threading.Event()
+    claim_deliveries = store.claim_deliveries
+    submitted: list[InboundMessage] = []
+
+    def blocked_claim_deliveries(*, limit: int = 20) -> list[TriggerDelivery]:
+        claim_started.set()
+        if not allow_claim.wait(timeout=2):
+            raise TimeoutError("test did not release delivery claim")
+        try:
+            return claim_deliveries(limit=limit)
+        finally:
+            claim_finished.set()
+
+    async def submit_turn(msg: InboundMessage) -> None:
+        submitted.append(msg)
+
+    monkeypatch.setattr(store, "claim_deliveries", blocked_claim_deliveries)
+    task = asyncio.create_task(
+        run_local_trigger_queue(
+            store=store,
+            submit_turn=submit_turn,
+            is_channel_enabled=_channel_is_enabled,
+            poll_interval_s=0.01,
+        )
+    )
+    try:
+        assert await asyncio.to_thread(claim_started.wait, 1)
+        task.cancel()
+        await asyncio.sleep(0)
+
+        assert not task.done()
+        assert len(list(store.inbox_dir.glob("*.json"))) == 1
+        assert list(store.processing_dir.glob("*.json")) == []
+
+        allow_claim.set()
+        with pytest.raises(asyncio.CancelledError):
+            await asyncio.wait_for(task, timeout=1)
+
+        assert claim_finished.is_set()
+        state_at_cancellation = (
+            sorted(path.name for path in store.inbox_dir.glob("*.json")),
+            sorted(path.name for path in store.processing_dir.glob("*.json")),
+        )
+        assert state_at_cancellation[0] == []
+        assert len(state_at_cancellation[1]) == 1
+        assert submitted == []
+
+        # The worker has returned before cancellation is visible, so no detached
+        # claim can move another file after the cancellation report.
+        await asyncio.sleep(0)
+        assert (
+            sorted(path.name for path in store.inbox_dir.glob("*.json")),
+            sorted(path.name for path in store.processing_dir.glob("*.json")),
+        ) == state_at_cancellation
+    finally:
+        allow_claim.set()
+        task.cancel()
+        with suppress(asyncio.CancelledError):
+            await task
+
+
+@pytest.mark.asyncio
+async def test_local_trigger_queue_cancellation_waits_for_startup_recovery(
+    tmp_path: Path,
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    store = LocalTriggerStore(tmp_path)
+    trigger = store.create(
+        name="PR review",
+        channel="websocket",
+        chat_id="chat-1",
+        session_key="websocket:chat-1",
+    )
+    store.enqueue(trigger.id, "Review PR #4591")
+    assert len(store.claim_deliveries()) == 1
+
+    restarted = LocalTriggerStore(tmp_path)
+    recovery_started = threading.Event()
+    allow_recovery = threading.Event()
+    recovery_finished = threading.Event()
+    recover_processing_deliveries = restarted.recover_processing_deliveries
+    submitted: list[InboundMessage] = []
+
+    def blocked_recovery() -> int:
+        recovery_started.set()
+        if not allow_recovery.wait(timeout=2):
+            raise TimeoutError("test did not release delivery recovery")
+        try:
+            return recover_processing_deliveries()
+        finally:
+            recovery_finished.set()
+
+    async def submit_turn(msg: InboundMessage) -> None:
+        submitted.append(msg)
+
+    monkeypatch.setattr(restarted, "recover_processing_deliveries", blocked_recovery)
+    task = asyncio.create_task(
+        run_local_trigger_queue(
+            store=restarted,
+            submit_turn=submit_turn,
+            is_channel_enabled=_channel_is_enabled,
+            poll_interval_s=0.01,
+        )
+    )
+    try:
+        assert await asyncio.to_thread(recovery_started.wait, 1)
+        task.cancel()
+        await asyncio.sleep(0)
+
+        assert not task.done()
+        assert list(restarted.inbox_dir.glob("*.json")) == []
+        assert len(list(restarted.processing_dir.glob("*.json"))) == 1
+
+        allow_recovery.set()
+        with pytest.raises(asyncio.CancelledError):
+            await asyncio.wait_for(task, timeout=1)
+
+        assert recovery_finished.is_set()
+        state_at_cancellation = (
+            sorted(path.name for path in restarted.inbox_dir.glob("*.json")),
+            sorted(path.name for path in restarted.processing_dir.glob("*.json")),
+        )
+        assert len(state_at_cancellation[0]) == 1
+        assert state_at_cancellation[1] == []
+        recovered_payload = json.loads(
+            (restarted.inbox_dir / state_at_cancellation[0][0]).read_text(encoding="utf-8")
+        )
+        assert recovered_payload["delivery"]["attempts"] == 1
+        assert submitted == []
+
+        await asyncio.sleep(0)
+        assert (
+            sorted(path.name for path in restarted.inbox_dir.glob("*.json")),
+            sorted(path.name for path in restarted.processing_dir.glob("*.json")),
+        ) == state_at_cancellation
+    finally:
+        allow_recovery.set()
+        task.cancel()
+        with suppress(asyncio.CancelledError):
+            await task
+
+
 @pytest.mark.asyncio
 async def test_local_trigger_queue_submits_bound_inbound_message(tmp_path: Path) -> None:
     store = LocalTriggerStore(tmp_path)
@@ -390,6 +549,7 @@ async def test_local_trigger_queue_rejects_unavailable_target_channel(tmp_path:
 @pytest.mark.asyncio
 async def test_local_trigger_queue_waits_for_submitted_turn_before_ack(
     tmp_path: Path,
+    monkeypatch: pytest.MonkeyPatch,
 ) -> None:
     store = LocalTriggerStore(tmp_path)
     trigger = store.create(
@@ -401,6 +561,17 @@ async def test_local_trigger_queue_waits_for_submitted_turn_before_ack(
     delivery = store.enqueue(trigger.id, "Review failed CI")
     submitted: list[InboundMessage] = []
     release = asyncio.Event()
+    completion_started = threading.Event()
+    allow_completion = threading.Event()
+    complete_delivery = store.complete_delivery
+
+    def _complete_delivery(claimed: TriggerDelivery) -> None:
+        completion_started.set()
+        if not allow_completion.wait(timeout=1):
+            raise TimeoutError("test did not release delivery completion")
+        complete_delivery(claimed)
+
+    monkeypatch.setattr(store, "complete_delivery", _complete_delivery)
 
     async def _submit_turn(msg: InboundMessage):
         submitted.append(msg)
@@ -430,6 +601,13 @@ async def test_local_trigger_queue_waits_for_submitted_turn_before_ack(
         assert stored.last_status is None
 
         release.set()
+        assert await asyncio.to_thread(completion_started.wait, 1)
+        stored = store.get(trigger.id)
+        assert stored is not None
+        assert stored.last_status is None
+        assert list(store.processing_dir.glob("*.json"))
+
+        allow_completion.set()
         for _ in range(100):
             stored = store.get(trigger.id)
             if stored and stored.last_status == "ok":
@@ -444,6 +622,7 @@ async def test_local_trigger_queue_waits_for_submitted_turn_before_ack(
         record = _read_run_record(store, delivery.id)
         assert record["status"] == "ok"
     finally:
+        allow_completion.set()
         task.cancel()
         with suppress(asyncio.CancelledError):
             await task
@@ -495,9 +674,115 @@ async def test_local_trigger_queue_requeues_when_submitted_turn_is_interrupted(
             await task
 
 
+@pytest.mark.asyncio
+async def test_accepted_turn_cancellation_settles_delivery_without_replay(
+    tmp_path: Path,
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    store = LocalTriggerStore(tmp_path)
+    trigger = store.create(
+        name="CI review",
+        channel="websocket",
+        chat_id="chat-1",
+        session_key="websocket:chat-1",
+    )
+    delivery = store.enqueue(trigger.id, "Review failed CI")
+    inbound: asyncio.Queue[InboundMessage] = asyncio.Queue()
+    work_accepted = asyncio.Event()
+    release_work = asyncio.Event()
+    effects: list[str] = []
+    publish_count = 0
+
+    async def publish(msg: InboundMessage) -> None:
+        nonlocal publish_count
+        publish_count += 1
+        await inbound.put(msg)
+
+    coordinator = LocalTriggerTurnCoordinator(
+        publish_inbound=publish,
+        dispatch=lambda _msg: asyncio.sleep(0),
+        is_running=lambda: True,
+    )
+
+    async def agent_worker() -> None:
+        msg = await inbound.get()
+        work_accepted.set()
+        await release_work.wait()
+        effects.append(msg.content)
+        coordinator.complete(msg)
+
+    agent_task = asyncio.create_task(agent_worker())
+    queue_task = asyncio.create_task(
+        run_local_trigger_queue(
+            store=store,
+            submit_turn=coordinator.submit,
+            is_channel_enabled=_channel_is_enabled,
+            poll_interval_s=0.01,
+        )
+    )
+    restarted_task: asyncio.Task[None] | None = None
+    try:
+        await asyncio.wait_for(work_accepted.wait(), timeout=1)
+        queue_task.cancel()
+        with pytest.raises(asyncio.CancelledError):
+            await asyncio.wait_for(queue_task, timeout=1)
+
+        # The delivery is durably settled before cancellation is reported even
+        # though the independently-owned agent turn has not completed yet.
+        assert effects == []
+        assert not list(store.processing_dir.glob("*.json"))
+        assert _read_run_record(store, delivery.id)["status"] == "accepted"
+
+        restarted = LocalTriggerStore(tmp_path)
+        replayed: list[InboundMessage] = []
+        poll_completed = threading.Event()
+        claim_deliveries = restarted.claim_deliveries
+
+        def tracked_claim_deliveries(*, limit: int = 20) -> list[TriggerDelivery]:
+            try:
+                return claim_deliveries(limit=limit)
+            finally:
+                poll_completed.set()
+
+        async def replay_submit(msg: InboundMessage) -> None:
+            replayed.append(msg)
+
+        monkeypatch.setattr(restarted, "claim_deliveries", tracked_claim_deliveries)
+        restarted_task = asyncio.create_task(
+            run_local_trigger_queue(
+                store=restarted,
+                submit_turn=replay_submit,
+                is_channel_enabled=_channel_is_enabled,
+                poll_interval_s=1,
+            )
+        )
+        assert await asyncio.to_thread(poll_completed.wait, 1)
+        assert replayed == []
+        assert restarted.claim_deliveries() == []
+
+        release_work.set()
+        await asyncio.wait_for(agent_task, timeout=1)
+        assert effects == ["Review failed CI"]
+        assert publish_count == 1
+    finally:
+        release_work.set()
+        if restarted_task is not None:
+            restarted_task.cancel()
+            with suppress(asyncio.CancelledError):
+                await restarted_task
+        queue_task.cancel()
+        with suppress(asyncio.CancelledError):
+            await queue_task
+        if not agent_task.done():
+            agent_task.cancel()
+        with suppress(asyncio.CancelledError):
+            await agent_task
+
+
 @pytest.mark.asyncio
 async def test_local_trigger_queue_does_not_retry_completed_agent_failure(
     tmp_path: Path,
+    monkeypatch: pytest.MonkeyPatch,
 ) -> None:
     store = LocalTriggerStore(tmp_path)
     trigger = store.create(
@@ -508,6 +793,17 @@ async def test_local_trigger_queue_does_not_retry_completed_agent_failure(
     )
     delivery = store.enqueue(trigger.id, "Review failed CI")
     started = asyncio.Event()
+    completion_started = threading.Event()
+    allow_completion = threading.Event()
+    complete_delivery = store.complete_delivery
+
+    def _complete_delivery(claimed: TriggerDelivery) -> None:
+        completion_started.set()
+        if not allow_completion.wait(timeout=1):
+            raise TimeoutError("test did not release delivery completion")
+        complete_delivery(claimed)
+
+    monkeypatch.setattr(store, "complete_delivery", _complete_delivery)
 
     async def _submit_turn(_msg: InboundMessage):
         started.set()
@@ -523,6 +819,13 @@ async def test_local_trigger_queue_does_not_retry_completed_agent_failure(
     )
     try:
         await asyncio.wait_for(started.wait(), timeout=1)
+        assert await asyncio.to_thread(completion_started.wait, 1)
+        stored = store.get(trigger.id)
+        assert stored is not None
+        assert stored.last_status is None
+        assert list(store.processing_dir.glob("*.json"))
+
+        allow_completion.set()
         for _ in range(100):
             stored = store.get(trigger.id)
             if stored and stored.last_status == "error":
@@ -540,6 +843,7 @@ async def test_local_trigger_queue_does_not_retry_completed_agent_failure(
         assert record["status"] == "error"
         assert record["error"] == "model failed"
     finally:
+        allow_completion.set()
         task.cancel()
         with suppress(asyncio.CancelledError):
             await task

tests/webui/test_forking.py

@@ -1,5 +1,7 @@
 from __future__ import annotations
 
+import asyncio
+import threading
 from types import SimpleNamespace
 from unittest.mock import ANY, AsyncMock, MagicMock
 
@@ -160,3 +162,50 @@ async def test_fork_handler_maps_invalid_source_and_internal_failure_to_stable_e
 
     channel.logger.warning.assert_called_once_with("fork_chat failed: {}", ANY)
     channel.send_webui_protocol_error.assert_awaited_once_with(connection, "fork_chat_failed")
+
+
+@pytest.mark.asyncio
+async def test_fork_cancellation_waits_for_creation_and_client_attachment(
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    started = threading.Event()
+    release = threading.Event()
+    connection = object()
+
+    def blocked_create(*_args, **_kwargs) -> tuple[str, str]:
+        started.set()
+        assert release.wait(timeout=1)
+        return "fork-id", "websocket:fork-id"
+
+    channel = SimpleNamespace(
+        send_webui_protocol_error=AsyncMock(),
+        attach_webui_fork=AsyncMock(),
+        gateway=SimpleNamespace(session_manager=object()),
+        logger=SimpleNamespace(warning=MagicMock()),
+    )
+    monkeypatch.setattr(forking, "create_webui_chat_fork", blocked_create)
+    task = asyncio.create_task(
+        forking.handle_webui_fork_chat(
+            channel,
+            connection,
+            {"source_chat_id": "source", "before_user_index": 0},
+        )
+    )
+    assert await asyncio.to_thread(started.wait, 1)
+    try:
+        task.cancel()
+        await asyncio.sleep(0)
+        assert not task.done()
+        channel.attach_webui_fork.assert_not_awaited()
+    finally:
+        release.set()
+
+    with pytest.raises(asyncio.CancelledError):
+        await asyncio.wait_for(task, timeout=1)
+
+    channel.attach_webui_fork.assert_awaited_once_with(
+        connection,
+        fork_id="fork-id",
+        fork_key="websocket:fork-id",
+    )
+    channel.send_webui_protocol_error.assert_not_awaited()

tests/webui/test_http_utils.py

@@ -2,8 +2,11 @@
 
 import gzip
 import json
+import time
+from types import SimpleNamespace
 
 from nanobot.webui.http_utils import http_json_response
+from nanobot.webui.ws_http import GatewayHTTPHandler
 
 
 def test_http_json_response_compresses_large_payload_when_gzip_is_accepted() -> None:
@@ -36,3 +39,29 @@ def test_http_json_response_does_not_compress_small_payload() -> None:
     assert "Content-Encoding" not in response.headers
     assert response.headers["Vary"] == "Accept-Encoding"
     assert json.loads(response.body) == payload
+
+
+def test_slow_http_log_records_route_scale_without_user_path_content() -> None:
+    records: list[str] = []
+
+    class _Logger:
+        def warning(self, message: str, *args: object) -> None:
+            records.append(message.format(*args))
+
+    handler = SimpleNamespace(_log=_Logger())
+    secret = "private-session-token"
+
+    GatewayHTTPHandler._log_slow_http(
+        handler,
+        f"/api/sessions/{secret}?query={secret}",
+        SimpleNamespace(status_code=200),
+        time.perf_counter() - 2,
+        input_chars=80,
+    )
+
+    assert len(records) == 1
+    assert "operation=/api/sessions" in records[0]
+    assert "status=200" in records[0]
+    assert "input_chars=80" in records[0]
+    assert "duration_ms=" in records[0]
+    assert secret not in records[0]

tests/webui/test_mcp_oauth_api.py

@@ -38,9 +38,12 @@ async def test_browser_flow_retries_current_server_and_ignores_unrelated_reload_
     received: dict[str, object] = {}
     reload_calls = 0
 
+    async def validate_url_target(_url: str) -> tuple[bool, str]:
+        return True, ""
+
     monkeypatch.setattr(
-        "nanobot.webui.mcp_oauth_api.validate_url_target",
+        "nanobot.webui.mcp_oauth_api.async_validate_url_target",
-        lambda _url: (True, ""),
+        validate_url_target,
     )
 
     async def connect(servers, _registry, *, oauth_handlers):
@@ -115,9 +118,12 @@ async def test_remote_http_flow_accepts_a_pasted_loopback_callback(
     connection = _Connection()
     received: dict[str, object] = {}
 
+    async def validate_url_target(_url: str) -> tuple[bool, str]:
+        return True, ""
+
     monkeypatch.setattr(
-        "nanobot.webui.mcp_oauth_api.validate_url_target",
+        "nanobot.webui.mcp_oauth_api.async_validate_url_target",
-        lambda _url: (True, ""),
+        validate_url_target,
     )
 
     async def connect(_servers, _registry, *, oauth_handlers):
@@ -187,9 +193,13 @@ async def test_browser_flow_surfaces_provider_denial_without_callback_descriptio
     monkeypatch: pytest.MonkeyPatch,
 ) -> None:
     manager = McpOAuthManager()
+
+    async def validate_url_target(_url: str) -> tuple[bool, str]:
+        return True, ""
+
     monkeypatch.setattr(
-        "nanobot.webui.mcp_oauth_api.validate_url_target",
+        "nanobot.webui.mcp_oauth_api.async_validate_url_target",
-        lambda _url: (True, ""),
+        validate_url_target,
     )
 
     async def connect(_servers, _registry, *, oauth_handlers):
@@ -233,9 +243,13 @@ async def test_browser_flow_blocks_unsafe_authorization_url(
     state: str,
 ) -> None:
     manager = McpOAuthManager()
+
+    async def validate_url_target(_url: str) -> tuple[bool, str]:
+        return url_is_safe, "private address"
+
     monkeypatch.setattr(
-        "nanobot.webui.mcp_oauth_api.validate_url_target",
+        "nanobot.webui.mcp_oauth_api.async_validate_url_target",
-        lambda _url: (url_is_safe, "private address"),
+        validate_url_target,
     )
 
     async def connect(_servers, _registry, *, oauth_handlers):

tests/webui/test_settings_routes.py

@@ -526,3 +526,179 @@ async def test_version_check_route_enforces_auth_and_bounds_failures(
     assert failed.status_code == 500
     assert json.loads(failed.body) == {"error": "version check failed"}
     assert "upstream secret body" not in failed.body.decode()
+
+
+@pytest.mark.asyncio
+async def test_mcp_oauth_start_cancellation_waits_for_config_and_flow_start(
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    config_started = threading.Event()
+    release_config = threading.Event()
+    start_entered = asyncio.Event()
+    release_start = asyncio.Event()
+    effects: list[str] = []
+    cfg = SimpleNamespace(
+        type="streamableHttp",
+        auth="oauth",
+        url="https://app.xmind.com/api/mcp",
+    )
+
+    def ensure_server(_query, *, config_path=None):
+        config_started.set()
+        assert release_config.wait(timeout=1)
+        effects.append("config_saved")
+        return "xmind", cfg
+
+    async def start(
+        name,
+        config,
+        redirect_uri,
+        *,
+        reload_mcp,
+        reset_credentials=False,
+    ):
+        assert (name, config, redirect_uri) == (
+            "xmind",
+            cfg,
+            "https://gateway.example/auth/mcp/callback",
+        )
+        assert callable(reload_mcp)
+        assert reset_credentials is False
+        effects.append("start_entered")
+        start_entered.set()
+        await release_start.wait()
+        effects.append("start_settled")
+        return {
+            "status": "authorization_required",
+            "flow_id": "flow-123",
+            "name": name,
+        }
+
+    monkeypatch.setattr(
+        "nanobot.webui.settings_routes.ensure_mcp_oauth_server",
+        ensure_server,
+    )
+    router = _router()
+    router._mcp_oauth = SimpleNamespace(start=start)
+    request = _mutation_request(
+        "/api/settings/mcp-oauth/start",
+        {"name": "xmind"},
+    )
+    task = asyncio.create_task(
+        router.dispatch(None, request, "/api/settings/mcp-oauth/start")
+    )
+
+    assert await asyncio.to_thread(config_started.wait, 1)
+    try:
+        task.cancel()
+        await asyncio.sleep(0)
+        assert not task.done()
+
+        release_config.set()
+        await asyncio.wait_for(start_entered.wait(), timeout=1)
+        assert effects == ["config_saved", "start_entered"]
+        assert not task.done()
+    finally:
+        release_config.set()
+        release_start.set()
+
+    with pytest.raises(asyncio.CancelledError):
+        await asyncio.wait_for(task, timeout=1)
+
+    assert effects == ["config_saved", "start_entered", "start_settled"]
+    await asyncio.sleep(0.05)
+    assert effects == ["config_saved", "start_entered", "start_settled"]
+
+
+@pytest.mark.asyncio
+async def test_model_settings_cancellation_waits_for_mutation_and_runtime_refresh(
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    mutation_started = threading.Event()
+    release_mutation = threading.Event()
+    effects: list[str] = []
+
+    def update(_query, *, config_path=None):
+        mutation_started.set()
+        assert release_mutation.wait(timeout=1)
+        effects.append("config_saved")
+        return {"updated": True}
+
+    def refresh_runtime_config() -> None:
+        effects.append("runtime_refreshed")
+
+    monkeypatch.setattr(
+        "nanobot.webui.settings_routes.update_agent_settings",
+        update,
+    )
+    path = "/api/settings/update"
+    request = _mutation_request(path, {"model_preset": "Codex"})
+    task = asyncio.create_task(
+        _router(refresh_runtime_config=refresh_runtime_config).dispatch(
+            None,
+            request,
+            path,
+        )
+    )
+
+    assert await asyncio.to_thread(mutation_started.wait, 1)
+    try:
+        task.cancel()
+        await asyncio.sleep(0)
+        assert not task.done()
+        assert effects == []
+    finally:
+        release_mutation.set()
+
+    with pytest.raises(asyncio.CancelledError):
+        await asyncio.wait_for(task, timeout=1)
+
+    assert effects == ["config_saved", "runtime_refreshed"]
+    await asyncio.sleep(0.05)
+    assert effects == ["config_saved", "runtime_refreshed"]
+
+
+@pytest.mark.asyncio
+async def test_image_settings_cancellation_waits_for_runtime_application(
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    reload_started = asyncio.Event()
+    release_reload = asyncio.Event()
+    effects: list[str] = []
+
+    def update(_query, *, config_path=None):
+        effects.append("config_saved")
+        return {"requires_restart": True}
+
+    async def reload_image(_bus):
+        effects.append("reload_started")
+        reload_started.set()
+        await release_reload.wait()
+        effects.append("reload_settled")
+        return {"ok": True, "requires_restart": False}
+
+    monkeypatch.setattr(
+        "nanobot.webui.settings_routes.update_image_generation_settings",
+        update,
+    )
+    monkeypatch.setattr(
+        "nanobot.webui.settings_routes.request_image_generation_reload",
+        reload_image,
+    )
+    path = "/api/settings/image-generation/update"
+    request = _mutation_request(path, {"enabled": True})
+    task = asyncio.create_task(_router().dispatch(None, request, path))
+
+    await asyncio.wait_for(reload_started.wait(), timeout=1)
+    assert effects == ["config_saved", "reload_started"]
+    task.cancel()
+    await asyncio.sleep(0)
+    assert not task.done()
+
+    release_reload.set()
+    with pytest.raises(asyncio.CancelledError):
+        await asyncio.wait_for(task, timeout=1)
+
+    assert effects == ["config_saved", "reload_started", "reload_settled"]
+    await asyncio.sleep(0.05)
+    assert effects == ["config_saved", "reload_started", "reload_settled"]