Prevent proactive compaction from archiving sessions that have an
in-flight agent task, avoiding mid-turn context truncation when a
task runs longer than the idle TTL.
Track text-only user messages that were flushed before the turn loop completes, then materialize an interrupted assistant placeholder on the next request so session history stays legal and later turns do not skip their own assistant reply.
Made-with: Cursor
Use session.add_message for the pre-turn user-message flush and add focused regression tests for crash-time persistence and duplicate-free successful saves.
Made-with: Cursor
The existing runtime_checkpoint mechanism preserves the in-flight
assistant/tool state if the process dies mid-turn, but the triggering
user message is only written to session history at the end of the turn
via _save_turn(). If the worker is killed (OOM, SIGKILL, a self-
triggered systemctl restart, container eviction, etc.) before the turn
completes, the user's message is silently lost: on restart, the session
log only shows the interrupted assistant turn without any record of
what the user asked. Any recovery tooling built on top of session logs
cannot reply because it has no prompt to reply to.
This patch appends the incoming user message to the session and flushes
it to disk immediately after the session is loaded and before the agent
loop runs, then adjusts the _save_turn skip offset so the final
persistence step does not duplicate it.
Limited to textual content (isinstance(msg.content, str)); list-shaped
content (media blocks) still flows through _save_turn's sanitization at
end of turn, preserving existing behavior for those cases.
* feat(agent): add mid-turn message injection for responsive follow-ups
Allow user messages sent during an active agent turn to be injected
into the running LLM context instead of being queued behind a
per-session lock. Inspired by Claude Code's mid-turn queue drain
mechanism (query.ts:1547-1643).
Key design decisions:
- Messages are injected as natural user messages between iterations,
no tool cancellation or special system prompt needed
- Two drain checkpoints: after tool execution and after final LLM
response ("last-mile" to prevent dropping late arrivals)
- Bounded by MAX_INJECTION_CYCLES (5) to prevent consuming the
iteration budget on rapid follow-ups
- had_injections flag bypasses _sent_in_turn suppression so follow-up
responses are always delivered
Closes#1609
* fix(agent): harden mid-turn injection with streaming fix, bounded queue, and message safety
- Fix streaming protocol violation: Checkpoint 2 now checks for injections
BEFORE calling on_stream_end, passing resuming=True when injections found
so streaming channels (Feishu) don't prematurely finalize the card
- Bound pending queue to maxsize=20 with QueueFull handling
- Add warning log when injection batch exceeds _MAX_INJECTIONS_PER_TURN
- Re-publish leftover queue messages to bus in _dispatch finally block to
prevent silent message loss on early exit (max_iterations, tool_error, cancel)
- Fix PEP 8 blank line before dataclass and logger.info indentation
- Add 12 new tests covering drain, checkpoints, cycle cap, queue routing,
cleanup, and leftover re-publish
Rename the auto compact module to autocompact.py for a cleaner path while keeping the AutoCompact type and behavior unchanged. Update the agent loop import to match.
When a user is idle for longer than a configured TTL, nanobot **proactively** compresses the session context into a summary. This reduces token cost and first-token latency when the user returns — instead of re-processing a long stale context with an expired KV cache, the model receives a compact summary and fresh input.
Each MCP server now connects in its own asyncio.Task to isolate anyio
cancel scopes and prevent 'exit cancel scope in different task' errors
when multiple servers (especially mixed transport types) are configured.
Changes:
- connect_mcp_servers() returns dict[str, AsyncExitStack] instead of None
- Each server runs in separate task via asyncio.gather()
- AgentLoop uses _mcp_stacks dict to track per-server stacks
- Tests updated to handle new API
Rename the boolean flag in _sanitize_persisted_blocks and alias the imported helper so session persistence cannot crash with TypeError when truncation is enabled.
Add allowed_env_keys config field to selectively forward host environment variables (e.g. GOPATH, JAVA_HOME) into the sandboxed subprocess environment, while keeping the default allow-list unchanged.
When the LLM returns an error (e.g. 429 quota exceeded, stream timeout),
streaming channels silently drop the error message because `_streamed=True`
is set in metadata even though no content was actually streamed.
This change:
- Skips setting `_streamed` when stop_reason is "error", so error messages
go through the normal channel.send() path and reach the user
- Stops appending error content to session history, preventing error
messages from polluting subsequent conversation context
- Exposes stop_reason from _run_agent_loop to enable the above check
Introduce a disabled_skills option in the config schema that allows
users to specify a list of skill names to be excluded. The setting is
threaded from config through Nanobot -> AgentLoop -> ContextBuilder ->
SkillsLoader. Disabled skills are filtered out from list_skills,
get_always_skills, and build_skills_summary. Four new test cases cover
the filtering behavior.
Add reraise parameter to AgentHook so hooks can opt out of exception
swallowing in CompositeHook._for_each_hook_safe. _LoopHook sets
reraise=True to let its exceptions propagate. _LoopHookChain is removed
and replaced with CompositeHook([loop_hook] + extra_hooks).
Signed-off-by: Lingao Meng <menglingao@xiaomi.com>
- Move _tool_hint implementation from loop.py to nanobot/utils/tool_hints.py
- Keep thin delegation in AgentLoop._tool_hint for backward compat
- Update test imports to test format_tool_hints directly
Made-with: Cursor
Add CI step to detect unused imports (F401) and unused variables (F841)
with ruff. Clean up existing violations:
- Remove unused Consolidator import in agent/__init__.py
- Remove unused re import in agent/loop.py
- Remove unused Path import in channels/feishu.py
- Remove unused ContentRepositoryConfigError import in channels/matrix.py
- Remove unused field and CommandHandler imports in channels/telegram.py
- Remove unused exception variable in channels/weixin.py
Replace single-stage MemoryConsolidator with a two-stage architecture:
- Consolidator: lightweight token-budget triggered summarization,
appends to HISTORY.md with cursor-based tracking
- Dream: cron-scheduled two-phase processor that analyzes HISTORY.md
and updates SOUL.md, USER.md, MEMORY.md via AgentRunner with
edit_file tools for surgical, fault-tolerant updates
New files: MemoryStore (pure file I/O), Dream class, DreamConfig,
/dream and /dream-log commands. 89 tests covering all components.
Introduce a CompositeHook that fans out lifecycle callbacks to an
ordered list of AgentHook instances with per-hook error isolation.
Extract the nested _LoopHook and _SubagentHook to module scope as
public LoopHook / SubagentHook so downstream users can subclass or
compose them. Add `hooks` parameter to AgentLoop.__init__ for
registering custom hooks at construction time.
Closes#2603
- Rename WebSearchConfig references to the new WebToolsConfig root struct that wraps both search config and global proxy settings
- Add 'enable' flag to WebToolsConfig to allow fully disabling all web-related tools (WebSearch, WebFetch) at runtime
- Update AgentLoop and SubagentManager to receive the full web config object instead of separate web_search_config/web_proxy parameters
- Update CLI command initialization to pass the consolidated web config struct instead of split fields
- Change default web search provider from brave to duckduckgo for better out-of-the-box usability (no API key required)
Read the default timezone from the agent context when wiring the cron tool so startup no longer depends on an out-of-scope local variable. Add a regression test to ensure AgentLoop passes the configured timezone through to cron.
Made-with: Cursor
Make cron use the configured agent timezone when a cron expression omits tz or a one-shot ISO time has no offset. This keeps runtime context, heartbeat, and scheduling aligned around the same notion of time.
Made-with: Cursor
Add agent-level timezone configuration with a UTC default, propagate it into runtime context and heartbeat prompts, and document valid IANA timezone usage in the README.
Replace the single _processing_lock (asyncio.Lock) with per-session locks
so that different sessions can process LLM requests concurrently, while
messages within the same session remain serialised.
An optional global concurrency cap is available via the
NANOBOT_MAX_CONCURRENT_REQUESTS env var (default 3, <=0 for unlimited).
Also re-binds tool context before each tool execution round to prevent
concurrent sessions from clobbering each other's routing info.
Tested in production and manually reviewed.
(cherry picked from commit c397bb4229e8c3b7f99acea7ffe4bea15e73e957)
