ReadFileTool had no file size check — reading a multi-GB file would
load everything into memory and crash the process. Now:
- Rejects files over ~512KB at the byte level (fast stat check)
- Truncates at 128K chars with a notice if content is too long
- Guides the agent to use exec with head/tail/grep for large files
This matches the protection already in ExecTool (10KB) and
WebFetchTool (50KB).
datetime.fromisoformat(at) raises ValueError for malformed strings,
which propagated uncaught and crashed the tool execution. Now catches
ValueError and returns a user-friendly error message instead.
When the LLM returns malformed tool arguments (e.g. a list or string
instead of a dict), validate_params would crash with AttributeError
in _validate() when calling val.items(). Now returns a clear
validation error instead of crashing.
When a cron job fires, the agent processes the scheduled message and
has access to the cron tool. If the original message resembles a
scheduling instruction (e.g. "remind me in 10 seconds"), the agent
would call cron.add again, creating an infinite feedback loop.
Add a cron-context flag to CronTool that blocks add operations during
cron job execution. The flag is set before process_direct() and cleared
in a finally block to ensure cleanup even on errors.
Fixes#1441
- Remove trailing whitespace and normalize blank lines
- Unify string quotes and line breaks for long lines
- Sort imports alphabetically across modules
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)
- SubagentManager tracks _session_tasks: session_key -> {task_id, ...}
- cancel_by_session() cancels all subagents for a session
- SpawnTool passes session_key through to SubagentManager
- /stop response reports subagent cancellation count
- Cleanup callback removes from both _running_tasks and _session_tasks
Builds on #1179
- Defer Brave API key resolution to execute() time instead of __init__,
so env var or config changes take effect without gateway restart
- Improve error message to reference actual config path
(tools.web.search.apiKey) instead of only mentioning env var
Fixes#1069 (issues 1 and 2 of 3)
Previously, WebSearchTool cached the API key in __init__, so keys added
to config.json or env vars after gateway startup were never picked up.
This caused a confusing 'BRAVE_API_KEY not configured' error even after
the key was correctly set (issue #1069).
Changes:
- Store the init-time key separately, resolve via property at each call
- Improve error message to guide users toward the correct fix
Closes#1069
Always provide an explicit httpx client to prevent MCP HTTP transport from inheriting httpx's default 5-second timeout, thereby avoiding conflicts with the upper layer tool's timeout settings.
`startswith` string comparison allows bypassing directory restrictions.
For example, `/home/user/workspace_evil` passes the check against
`/home/user/workspace` because the string starts with the allowed path.
Replace with `Path.relative_to()` which correctly validates that the
resolved path is actually inside the allowed directory tree.
Fixes#888
Uses difflib to find the best match and shows a helpful diff,
making it easier to debug edit_file failures.
Co-authored-by: Cursor <cursoragent@cursor.com>