MCP Server Python

🚀 MCP Server Using Python

This project demonstrates an MCP server built with Python, enabling seamless integration of custom tools into AI - assisted development environments.

🚀 Quick Start

The Model Context Protocol (MCP) is a framework that allows you to integrate custom tools into AI - assisted development environments, such as Cursor AI. An MCP server exposes functionalities (e.g., data retrieval or code analysis) so that LLM - based IDEs can call these tools on demand.

This project showcases an MCP server built with Python, offering two basic functions. One function, add, accepts two numbers and returns their sum. The other function, getApiKey, retrieves an API key from the environment (via the API_KEY variable). For more information, refer to Introduction to the Model Context Protocol.

✨ Features

• MCP Integration: Provide functions to LLM - based IDEs.
• Addition Tool: Accept two numeric parameters and return their sum.
• Environment Variable Retrieval: Demonstrate how to load example environment variables from a configuration file.
• CLI Support: Offer a command - line interface through mcp[cli] for easy local development and testing.
• Standard I/O Transport: Integrate with the development environment using standard I/O for seamless communication.

📦 Installation

1. Clone the Repository

   git clone <repository_url>
   cd <repository_directory>
   

2. Set Up a Virtual Environment Create and activate a Python virtual environment:

   python -m venv .venv
   

   On Linux and macOS:

   source .venv/bin/activate  # or .venv/bin/activate
   

   On Windows:

   .venv\Scripts\activate
   

3. Install Dependencies Use the following command to install the project's required dependencies:

   pip install -r requirements.txt
   

📚 Documentation

Configure the MCP Server

1. Initialize the MCP Server In a Python script, initialize the MCP server using the MCP library and connect via standard I/O transport.
2. Define Tools
   • add: Define it to accept two numeric inputs and return their sum.
   • getApiKey: Retrieve the API key from the environment variable API_KEY and return it.

Test the MCP Server

1. Start the MCP Server Run the following command in the terminal to start the MCP server:

   python server.py
   

2. Call Tools Use the following command to test the add tool:

   echo "add 3 5" | nc -U ipc://localhost/mcp-server
   

   Use the following command to test the getApiKey tool:

   echo "getApiKey" | nc -U ipc://localhost/mcp-server
   

Integrate into Cursor AI

1. Configure the mcp.json File Create an mcp.json file in the project root directory and add the following content:

   {
     "servers": [
       {
         "name": "local - mcp - server",
         "transport": "ipc",
         "endpoint": "ipc://localhost/mcp - server"
       }
     ],
     "tools": [
       {
         "name": "add",
         "description": "Calculate the sum of two numbers",
         "parameters": [
           {"name": "a", "type": "number"},
           {"name": "b", "type": "number"}
         ]
       },
       {
         "name": "getApiKey",
         "description": "Get the API key",
         "parameters": []
       }
     ]
   }
   

2. Start Cursor AI In Cursor, navigate to the settings and enable MCP support. Configure local - mcp - server as the default server.
3. Use Tools In Cursor's console, enter the following commands:

   add 3 and 5
   

   Or

   what is my API key?
   

   The AI agent will infer the add or getApiKey tool from your MCP server and execute accordingly.

📄 License

This project is licensed under the MIT License.

🔗 References

• Model Context Protocol: Python SDK
• [How to Build MCP Servers and Clients from Scratch](https://dev.to/composiodev/how - to - build - mcp - servers - and - clients - from - scratch - 4o2f)
• Introduction to the Model Context Protocol