- Add separate retry budget for InvalidStreamError (NO_FINISH_REASON/NO_RESPONSE_TEXT)
- Fix hasFinishReason logic to prevent usage-only chunks from overwriting existing finish reason
- Update tests to use fake timers for retry delay handling
- Add test for trailing usage-only chunk handling
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
After the merged finish+usage response was yielded, handleChunkMerging's
hasPendingFinish flag stayed true due to reference-shared candidates.
Trailing empty chunks would trigger another merge, yielding the same
functionCall parts again and causing duplicate tool executions in the UI.
Add a finishYielded guard in processStreamWithLogging so that once the
merged finish response is yielded, all subsequent chunks skip merging
entirely and only absorb late-arriving usage metadata.
Closes#2121
Consolidate individual message component files into two modules:
- ConversationMessages.tsx for user/assistant/think messages
- StatusMessages.tsx for info/warning/error/retry messages
Rename Gemini* components to Assistant*/Think* for clarity.
Add 'code' semantic color token for consistent code styling.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
Add four new custom commands for Qwen Code CLI:
- code-review: Review pull requests with detailed analysis
- commit: Generate commit messages and push changes
- create-issue: Draft and submit GitHub issues
- create-pr: Create well-structured pull requests
These commands provide structured workflows for common GitHub operations.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
Add ability to capture multiple screenshots at intervals during
long-running terminal output (e.g., progress bars). Optionally
generates animated GIFs from captured frames using ffmpeg.
Features:
- Streaming capture at configurable intervals
- Early stop when output stabilizes (3 consecutive unchanged frames)
- Duplicate frame skipping
- Animated GIF generation via ffmpeg concat demuxer
- Auto-cleanup of output directory before each run
- Configurable delay before starting captures
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
Previously, only countdown-based errors were cleared when user starts
a new query. Static errors (like "Press Ctrl+Y to retry") remained
visible. Now both types of errors are properly cleared.
- Add /context command to the Interface and Workspace Control table
- Document the context window usage breakdown feature
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
### Shell & Interactive Terminal Improvements
- PTY shell is now enabled by default instead of disabled
- Improved shell output rendering, process termination, and added fallback warning
- Background commands now properly capture subprocess PIDs on non-Windows
### Coding Plan Improvements
- Simplified auth message, added /model tip, improved system info display
- Reordered model list to prioritize glm-5, kimi-k2.5, MiniMax-M2.5
- Model selection is now preserved when updating if the model still exists
### Other Changes
- Added shared symlink utility; debug logs now have latest alias
- Unknown settings warnings go to debug log instead of user-facing warnings
- Fixed subagent confirmation state detection
- Removed debug UI from AgentCreationWizard
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
The LLM-based loop detection was causing issues where the process would
stop unexpectedly. This change removes the LLM-based loop detection
feature entirely and sets skipLoopDetection to true by default to
prevent false positives from interrupting user sessions.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
The colorizeCode function signature was updated to include a tabWidth
parameter (defaulting to 4). This updates the test assertions to match
the new function signature.
Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
