DATE:
AUTHOR:
The LangChain Team
LangChain

Middleware in LangChain 1.0 alpha

DATE:
AUTHOR: The LangChain Team

Middleware in LangChain 1.0 gives you fine-grained control around the core agent loop of models + tools. It includes:

  • Human-in-the-loop so you can approve, modify, or reject risky tool calls before execution

  • Summarization of older messages as you approach token limits, so you can manage conversation history

Read the blog about middleware and try out middleware in the LangChain 1.0 alpha here.

Powered by LaunchNotes