langchain
https://github.com/langchain-ai/langchain
Python
⚡ Building applications with LLMs through composability ⚡
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Python not yet supported40 Subscribers
Add a CodeTriage badge to langchain
Help out
- Issues
- Streamed refusals are dropped, so structured output raises a generic `ValueError` instead of `OpenAIRefusalError`
- [langchain-deepseek] ChatDeepSeek raises KeyError('messages') with use_responses_api=True
- Tool schemas and invocation diverge for valid Pydantic and Python signatures
- fix(openrouter): close `httpx` clients created for attribution headers
- ShellToolMiddleware leaks subprocess and temp directory when agent is interrupted and never reaches after_agent
- `ChatAnthropic(stream_usage=False)` also drops `stop_reason`, `stop_sequence` and `model_name` from `response_metadata`, not just usage metadata
- Bug: count_tokens_approximately inflates estimates 5-10x for reasoning/thinking blocks
- ChatAnthropic streaming on thinking-by-default models (claude-opus-5/sonnet-5) yields malformed content; replaying it 400s the next turn
- dumps()/dumpd() raise TypeError for any dict that has an "lc" key alongside a non-JSON-native value (e.g. datetime)
- AsyncCallbackManager.on_llm_start drops non-inline handlers when an inline handler is registered
- Docs
- Python not yet supported