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 supported39 Subscribers
Add a CodeTriage badge to langchain
Help out
- Issues
- Add `BackendIntegrationTests` for Deep Agents `BackendProtocol` implementations
- create_agent discards state_schema subclass in inferred return type
- convert_to_openai_messages mutates the caller's input message in place
- feat(agents): `AgentRuntime` fields + `_build_runtime` hook for middleware
- AgentExecutor._execute_model_async uses ainvoke (non-streaming) — on_llm_new_token never fires; LangGraph use_astream=True does not enable token-level streaming
- fix(langchain): sanitize anthropic cache markers on fallback retries
- Agent factory _execute_model_async uses ainvoke (non-streaming) and cannot trigger on_llm_new_token callback; LangGraph use_astream=True only controls graph-level streaming not model token-by-token streaming
- Importing langchain.agents breaks BaseLLM class construction on Pydantic main
- Responses API: parsed_arguments leaks from ParsedResponseFunctionToolCall into AIMessage content, breaking the next turn
- ShellSession._collect_output times out when command output lacks a trailing newline
- Docs
- Python not yet supported