Langchain MCP Tools Py Usage

What is langchain-mcp-tools-py-usage

langchain-mcp-tools-py-usage is a client tool designed to demonstrate the integration of the Model Context Protocol (MCP) server tools with the LangChain ReAct Agent. Developed by Hideya, this tool provides a practical way to explore how AI systems, like those powered by Anthropic's Claude, can securely connect with various data sources and tools using MCP. By leveraging the convert_mcp_to_langchain_tools() utility function from the langchain_mcp_tools package, this client facilitates the parallel initialization of multiple MCP servers and converts their tools into LangChain-compatible formats. This setup empowers AI clients to access and utilize real-time information and data sources seamlessly.

How to Use langchain-mcp-tools-py-usage

To get started with langchain-mcp-tools-py-usage, follow these steps:

1. Install Dependencies:

   • Use the command:

     make install
     

   This command will set up all necessary dependencies for the application.

2. Set Up API Key:

   • Copy the environment template file and configure your API credentials:

     cp .env.template .env
     

   • Edit the .env file to include your API key from Anthropic or OpenAI. Note that the .env file is set to be ignored by version control to protect your credentials.

3. Run the Application:

   • Start the application using:

     make start
     

   Note that the initial run might take some time as it sets up the necessary configurations.

This client requires Python 3.11 or higher. Optionally, you can install uv for Python-based MCP servers or npm for Node.js-based MCP servers to expand the functionality.

Key Features of langchain-mcp-tools-py-usage

• MCP Integration: Demonstrates the integration of MCP server tools with LangChain ReAct Agent for a seamless AI and data interaction experience.

• Utility Functionality: Utilizes the convert_mcp_to_langchain_tools() function to efficiently initialize and convert multiple MCP servers' tools into a LangChain-compatible format.

• Multiple AI Model Support: Primarily uses Anthropic's claude-3-7-sonnet-latest as the language model, with optional support for OpenAI's o3-mini included in the codebase.

• Secure and Versatile: By using the Model Context Protocol, it ensures secure connections to various data sources, supporting a wide range of real-time data access.

• Cross-Platform Compatibility: While the primary focus is on Python, a TypeScript equivalent of this client is also available, offering flexibility in development environments.

This tool is ideal for developers and AI enthusiasts looking to explore the practical applications of MCP in enhancing AI capabilities through secure and efficient data access.