🔌 MCP Toolkit: Multi-Server AI Agent with LangChain

Build intelligent agents that seamlessly connect to multiple MCP servers for math operations and real-time data retrieval.

Overview

This project sets up two MCP servers:

• Math Server - Provides arithmetic tools (add, multiply, divide) via stdio transport
• Weather Server - Provides weather lookup via streamable HTTP transport

A LangChain agent connects to both servers and can use their tools to answer questions.

Setup

Prerequisites

• Python 3.13+
• uv package manager

Installation

uv sync

Environment Variables

Create a .env file with your Groq API key:

GROQ_API_KEY=gsk_your_api_key_here

Get your API key from console.groq.com

Usage

1. Start the Weather Server

In one terminal:

uv run python weather.py

This starts an HTTP server on http://127.0.0.1:8000

2. Run the Client

In another terminal:

uv run python client.py

The client will:

1. Connect to both MCP servers (math via stdio, weather via HTTP)
2. Create a ReAct agent with access to all tools
3. Execute sample queries for math and weather

Project Structure

├── client.py        # LangChain agent that connects to MCP servers
├── mathserver.py    # MCP server with arithmetic tools (stdio)
├── weather.py       # MCP server with weather tool (HTTP)
├── requirements.txt # Python dependencies
├── pyproject.toml   # Project configuration
└── .env             # Environment variables (not committed)

Available Tools

Math Server (stdio)

• add(a, b) - Add two numbers
• multiply(a, b) - Multiply two numbers
• divide(a, b) - Divide two numbers

Weather Server (HTTP)

• get_weather(location) - Get weather for a location

Dependencies

• langchain-groq - Groq LLM integration
• langchain-mcp-adapters - MCP client adapters for LangChain
• mcp - Model Context Protocol SDK
• langgraph - LangGraph for agent orchestration