[Release] langchain-ai/langchain langchain==1.3.15: langchain==1.3.15

LangChain v1.3.15 release includes 30+ improvements focused on middleware enhancements, bug fixes, and dependency updates. Key additions include trace_policy exposure on AgentMiddleware, state_schema parameter for wrap_tool_call, and LangSmith provider integration in init_chat_model. The release addresses critical issues with history preservation, structured output handling, HITL approval gates, and tool call management while maintaining security through dependency bumps.

Key Points

• •Expose trace_policy on AgentMiddleware for better tracing control and observability
• •Add state_schema parameter to wrap_tool_call for improved state management in tool interactions
• •Integrate LangSmith provider into init_chat_model for seamless model initialization and monitoring
• •Fix SummarizationMiddleware to preserve conversation history on summary failures
• •Prevent orphaned tool_calls in ToolCallLimitMiddleware end behavior to ensure clean state transitions
• •Handle malformed structured-output responses gracefully to improve robustness
• •Stop HITL approval gates from silently failing open—enforce explicit approval requirements
• •Filter internal middleware model calls from messages projection to reduce noise in message history
• •Add reasoning_effort as standard chat model parameter for enhanced model control
• •Clear stale structured_response between checkpointed turns to prevent state contamination

Workflow Diagram

Concepts