MCP.so
ログイン

LangChain Agent with MCP Servers

@esakrissa

リポジトリ

LangChain Agent with MCP Servers について

LangChain Agent with MCP Servers: Using LangChain MCP Adapters for tool integration.

#langchain#mcp#tool-integration

基本情報

カテゴリ

AI とエージェント

ランタイム

python

トランスポート

stdio

公開者

esakrissa

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-esakrissa": {
      "command": "python",
      "args": [
        "-m",
        "venv",
        ".venv"
      ]
    }
  }
}

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 Agent with MCP Servers?

A LangChain agent that uses the Model Context Protocol (MCP) to integrate with external services such as Tavily Search, a mock weather service, and a math expression evaluator. It is built for developers who want a modular, LangGraph-powered agent with graceful shutdown and easy extensibility.

How to use LangChain Agent with MCP Servers?

Install dependencies, create a virtual environment, and add API keys for OpenAI and Tavily to a .env file. Run python src/agent.py from the command line and enter a query when prompted. The agent automatically selects and invokes the appropriate MCP‑server tool.

Key features of LangChain Agent with MCP Servers

  • Integrates with Tavily Search, weather, and math MCP servers
  • Uses LangGraph’s ReAct agent pattern for dynamic tool selection
  • Graceful shutdown via signal handling and subprocess tracking
  • Robust error handling throughout the application
  • Modular design makes adding new MCP servers straightforward

Use cases of LangChain Agent with MCP Servers

  • Perform web searches and retrieve news via Tavily
  • Get mock weather information for any location
  • Evaluate mathematical expressions
  • Demonstrate a LangChain agent that dynamically calls external MCP tools

FAQ from LangChain Agent with MCP Servers

What runtime dependencies are required?

Python 3.8+, a virtual environment, and the dependencies listed in setup.py. API keys for OpenAI and Tavily must be stored in a .env file.

How can I add a new MCP server?

Create a new file in src/mcpserver/ with proper signal handling, expose it in src/mcpserver/__init__.py, and add its configuration in src/agent.py.

What transport or authentication does the agent use?

MCP servers run as subprocesses and communicate via stdin/stdout. Authentication is handled via API keys passed in the .env file.

Does the agent support graceful shutdown?

Yes. It captures SIGINT and SIGTERM, tracks all child processes, and ensures they terminate cleanly. A shutdown flag prevents new operations during teardown, and asyncio allows in‑progress tasks to finish when possible.

What services are available out‑of‑the-box?

Tavily Search (web and news), a mock weather server, and a math expression evaluator.

コメント

「AI とエージェント」の他のコンテンツ

MCP-LLM Bridge

patruff

Bridge between Ollama and MCP servers, enabling local LLMs to use Model Context Protocol tools

976

Shell and Coding agent for Claude and other mcp clients

rusiaaman

Shell and coding agent on mcp clients

666

Sequential Thinking Multi-Agent System (MAS)

FradSer

An advanced sequential thinking process using a Multi-Agent System (MAS) built with the Agno framework and served via MCP.

306

MCP Claude Code

SDGLBL

MCP implementation of Claude Code capabilities and more

302

Solon Ai

opensolon

Java AI application development framework (supports LLM-tool,skill; RAG; MCP; Agent-ReAct,Team-Agent). Compatible with java8 ~ java25. It can also be embedded in SpringBoot, jFinal, Vert.x, Quarkus, and other frameworks.

416