Langchain MCP Tools Ts Usage

What is langchain-mcp-tools-ts-usage

The langchain-mcp-tools-ts-usage is a simple Model Context Protocol (MCP) client that demonstrates how to use MCP server tools through the LangChain ReAct Agent. Developed by the company hideya, this server leverages a utility function called convertMcpToLangchainTools() from the package @h1deya/langchain-mcp-tools. This function facilitates the parallel initialization of multiple MCP servers and transforms their available tools into an array of LangChain-compatible tools.

The server uses Anthropic's claude-3-5-sonnet-latest as the primary language model, but it also includes setup code for OpenAI's gpt-4o, although this is commented out by default. Essentially, langchain-mcp-tools-ts-usage acts as a bridge, enabling AI systems to utilize external data sources and tools in a standardized manner through MCP.

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

To get started with langchain-mcp-tools-ts-usage, you will need to meet certain prerequisites and follow a straightforward setup process.

Prerequisites

• Node.js 16+: Ensure you have Node.js version 16 or higher.
• npm 7+: Required to run Node.js-based MCP servers.
• API Key: Obtain an API key from Anthropic (or optionally from OpenAI if using gpt-4o).

Optional

• uv (uvx): Needed if you wish to run Python-based MCP servers.

Setup and Running the Server

1. Install Dependencies: Open your terminal and run the following command to install necessary dependencies:

   npm install
   

2. Setup API Key: Create an .env file by copying the template and update it with your API key:

   cp .env.template .env
   

   Ensure your API key is correctly placed in this file, which is ignored by version control to protect your credentials.

3. Run the Application: Finally, start the application by running:

   npm start
   

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

• Easy Integration with MCP: The server provides a seamless way to connect AI systems with MCP tools, allowing for secure and efficient access to external data sources.

• Multi-Model Support: By default, it uses Anthropic's claude-3-5-sonnet-latest but is also compatible with OpenAI's models, providing flexibility depending on your requirements.

• Efficient Initialization: The utility function convertMcpToLangchainTools() supports the parallel initialization of multiple MCP servers, optimizing performance and reducing setup complexity.

• LangChain Compatibility: Tools from MCP servers are converted into LangChain-compatible formats, making it easy to incorporate them into existing AI workflows.

• Secure Handling of Credentials: The server setup includes a .env file that ensures API keys and sensitive information remain secure and are excluded from version control.

The langchain-mcp-tools-ts-usage is ideal for developers looking to integrate AI systems with MCP-supported data and tools, providing a robust and standardized approach to accessing and utilizing external resources.