Add a built-in tool that lets the agent inspect and modify its own
runtime state (model, iterations, context window, etc.).
Key features:
- inspect: view current config, usage stats, and subagent status
- modify: adjust parameters at runtime (protected by type/range validation)
- Subagent observability: inspect running subagent tasks (phase,
iteration, tool events, errors) — subagents are no longer a black box
- Watchdog corrects out-of-bounds values on each iteration
- Enabled by default in read-only mode (self_modify: false)
- All changes are in-memory only; restart restores defaults
- Comprehensive test suite (90 tests)
Includes a self-awareness skill (always-on) with progressive disclosure:
SKILL.md for core rules, references/examples.md for detailed scenarios.
PyJWT and cryptography are optional msteams deps; they should not be
bundled into the generic dev install. Tests now skip the entire file
when the deps are missing, following the dingtalk pattern.
- Migrate "after tools" inline drain to use _try_drain_injections,
completing the refactoring (all 6 drain sites now use the helper).
- Move checkpoint emission into _try_drain_injections via optional
iteration parameter, eliminating the leaky split between helper
and caller for the final-response path.
- Extract _make_injection_callback() test helper to replace 7
identical inject_cb function bodies.
- Add test_injection_cycle_cap_on_error_path to verify the cycle
cap is enforced on error exit paths.
When the agent runner exits due to LLM error, tool error, empty response,
or max_iterations, it breaks out of the iteration loop without draining
the pending injection queue. This causes leftover messages to be
re-published as independent inbound messages, resulting in duplicate or
confusing replies to the user.
Extract the injection drain logic into a `_try_drain_injections` helper
and call it before each break in the error/edge-case paths. If injections
are found, continue the loop instead of breaking. For max_iterations
(where the loop is exhausted), drain injections to prevent re-publish
without continuing.
Instead of a separate skill discovery system, extend Dream's two-phase
pipeline to also detect reusable behavioral patterns from conversation
history and generate SKILL.md files.
Phase 1 gains a [SKILL] output type for pattern detection.
Phase 2 gains write_file (scoped to skills/) and read access to builtin
skills, enabling it to check for duplicates and follow skill-creator's
format conventions before creating new skills.
Inspired by PR #3039 by @wanghesong2019.
Co-authored-by: wanghesong2019 <wanghesong2019@users.noreply.github.com>
The plain reply() uses cmd="reply" which does not support "text" msgtype
and causes WeCom API to return errcode=40008 (invalid message type).
Unify both progress and final text messages to use reply_stream()
(cmd="aibot_respond_msg"), differentiating via finish flag.
Fixes#2999
Dream Phase 2 uses fail_on_tool_error=True, which terminates the entire
run on the first tool error (e.g. old_text not found in edit_file).
Normal agent runs default to False so the LLM can self-correct and retry.
Dream should behave the same way.
Feishu's GetMessageResource API only accepts 'image' or 'file' as the
type parameter. Video messages have msg_type='media', which was passed
through unchanged, causing error 234001 (Invalid request param). Now
both 'audio' and 'media' are converted to 'file' for download.
Remove the fragile barrier-based synchronization that could cause
deadlock when the second request is scheduled first. Instead, rely
on the session lock for serialization and handle either execution
order in assertions.
--workspace and --config now work as initial defaults in interactive mode:
- The wizard starts with these values pre-filled
- Users can view and modify them in the wizard
- Final saved config reflects user's choices
This makes the CLI args more useful for interactive sessions while
still allowing full customization through the wizard.
Tests for non-interactive onboard mode now explicitly use --no-interactive
flag since the default changed to interactive mode.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add `_SELECT_FIELD_HINTS` for select fields with predefined choices
(e.g., reasoning_effort: low/medium/high with hint text)
- Add `_select_with_back()` using prompt_toolkit for custom key bindings
- Support Escape and Left arrow keys to go back in menus
- Apply to field config, provider selection, and channel selection menus
- Add model_info.py module with litellm-based model lookup
- Provide autocomplete suggestions for model names
- Auto-fill context_window_tokens when model changes (only at default)
- Add "Get recommended value" option for manual context lookup
- Dynamically load provider keywords from registry (no hardcoding)
Resolves#2018
- Add CONTRIBUTING.md with detailed contribution guidelines
- Add branching strategy section to README.md explaining main/nightly branches
- Include maintainer information and development setup instructions
- test_channel_plugins: fix assertion logic for discoverable channels
- test_filesystem_tools: normalize path separators for Windows
- test_tool_validation: use python to generate output, avoid cmd line limits
- test_context_prompt_cache: Update test to reflect merged runtime
context and user message (commit ad99d5a merged them into one)
- Remove test_cron_commands.py: cron add CLI command was removed
in commit c05cb2e (unified scheduling via cron tool)
- test_context_prompt_cache: Update test to reflect merged runtime
context and user message (commit ad99d5a merged them into one)
- Remove test_cron_commands.py: cron add CLI command was removed
in commit c05cb2e (unified scheduling via cron tool)
- Add --workspace/-w flag to specify workspace directory
- Add --config/-c flag to specify config file path
- Move cron store to workspace directory for per-instance isolation
- Enable running multiple nanobot instances simultaneously
When allow_from is not configured, block all access by default
instead of allowing everyone. This prevents unauthorized access
when channels are enabled without explicit allow lists.
- Add handler for SIGTERM to prevent "Terminated" message on Linux
- Add handler for SIGHUP for terminal closure handling
- Ignore SIGPIPE to prevent silent process termination
- Change os._exit(0) to sys.exit(0) for proper cleanup
Fixes issue #1365
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
A refactoring in commit 132807a introduced a regression where the final
response was silently discarded whenever the message tool was used,
regardless of the target. This restored the original logic from PR #832
that only suppresses the final reply when the message tool sends to the
same (channel, chat_id) as the original message.
Changes:
- message.py: Replace _sent_in_turn: bool with _turn_sends: list[tuple]
to track actual send targets, add get_turn_sends() method
- loop.py: Check if (msg.channel, msg.chat_id) is in sent_targets before
suppressing final reply. Also move the "Response to" log after the
suppress check to avoid misleading logs.
- Add unit tests for the suppress logic
This ensures:
- Email sent via message tool → Feishu still gets confirmation
- Message tool sends to same Feishu chat → No duplicate (suppressed)
Previously, `autoReplyEnabled=false` would block ALL email sends,
including proactive emails triggered from other channels (e.g., asking
nanobot on Feishu to send an email).
Now `autoReplyEnabled` only controls automatic replies to incoming
emails, not proactive sends. This allows users to disable auto-replies
while still being able to ask nanobot to send emails on demand.
Changes:
- Check if recipient is in `_last_subject_by_chat` to determine if
it's a reply
- Only skip sending when it's a reply AND auto_reply_enabled is false
- Add test for proactive send with auto_reply_enabled=false
- Update existing test to verify reply behavior
chengyongruandXubin Ren
