MCP.so
ログイン

langchain-mcp

@rectalogic

langchain-mcp について

Model Context Protocol tool support for LangChain

基本情報

カテゴリ

AI とエージェント

ライセンス

MIT

ランタイム

python

トランスポート

stdio

公開者

rectalogic

設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "langchain-mcp": {
      "command": "uv",
      "args": [
        "run",
        "tests/demo.py",
        "Read and summarize the file ./LICENSE"
      ]
    }
  }
}

ツール

ツールは検出されませんでした

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is langchain-mcp?

langchain-mcp provides Model Context Protocol (MCP) tool calling support within LangChain. It enables LangChain applications to use tools exposed by any MCP server by wrapping them as LangChain BaseTool instances. This is for developers building LangChain-based agents or chains that need to call external tools via the MCP standard.

How to use langchain-mcp?

Create a langchain_mcp.MCPToolkit by passing it an mcp.ClientSession, then call await toolkit.initialize() followed by toolkit.get_tools() to retrieve a list of LangChain BaseTools. The toolkit handles the MCP client‑server lifecycle and tool discovery.

Key features of langchain-mcp

  • Integrates MCP tool calling into LangChain workflows
  • Exposes an async MCPToolkit class for session management
  • Automatically converts MCP tools into BaseTool objects
  • Works with any MCP server that communicates over a ClientSession

Use cases of langchain-mcp

  • Adding filesystem, database, or API tools from MCP servers to a LangChain agent
  • Combining MCP tools with LangChain’s chat models and chains in a unified pipeline

FAQ from langchain-mcp

Is this the official LangChain MCP integration?

No. LangChain now has a more official implementation called langchain-mcp-adapters. This repository is a community alternative.

What does langchain-mcp require?

It requires Python and can be installed via PyPI as langchain-mcp. The demo also uses the uv tool for running.

How do I run the included demo?

Set the GROQ_API_KEY environment variable and execute uv run tests/demo.py "Read and summarize the file ./LICENSE" (or equivalent) against an MCP filesystem server.

Does langchain-mcp support authentication or transport customization?

コメント

「AI とエージェント」の他のコンテンツ