MCP Template

🚀 MCP Hello, World! Example

This project demonstrates a simple implementation of the Model Context Protocol (MCP) using a Flask server integrated with OpenAI. It shows how to create an MCP server that provides a number addition function and how to use this function via OpenAI's function calling capabilities.

🚀 Quick Start

This example project showcases the integration of the Model Context Protocol (MCP) with OpenAI. It allows you to perform simple arithmetic operations through OpenAI's function calling feature.

✨ Features

Server (server.py)

• A Flask-based MCP server that implements the following features:
  • A simple number addition function (add_numbers).
  • MCP protocol endpoints for function listing and execution.
  • JSON Schema definition for the addition function.
  • Runs on http://localhost:8080.

Client (client.py)

• Integrates the OpenAI API with the MCP server.
• Demonstrates the following:
  • Retrieving available functions from the MCP server.
  • Converting MCP functions to OpenAI function specifications.
  • Making function calls via OpenAI's chat completion.
  • Processing and displaying the results.

📦 Installation

Prerequisites

• Python 3.x
• OpenAI API key

Steps

1. Clone the repository.
2. Install the dependencies:

pip install -r requirements.txt

📚 Documentation

Configuration

1. Set your OpenAI API key:

export OPENAI_API_KEY="your-api-key-here"

Usage

1. Start the MCP server:

python server.py

2. Run the client in another terminal:

python client.py

The client will perform the following actions:

1. Connect to the MCP server.
2. Retrieve available functions.
3. Start a conversation with OpenAI's model.
4. Demonstrate addition via function call (e.g., calculate the sum of 42 and 17).
5. Display the results.

Project Structure

.
├── README.md           # Project documentation
├── requirements.txt    # Python dependencies
├── server.py          # MCP server implementation
└── client.py          # OpenAI client implementation

Dependencies

• python-dotenv: Environment variable management.
• openai: OpenAI API client.
• flask: Web server framework.
• requests: HTTP client library.

Example Output

When running the client, you will see:

1. A list of available MCP functions.
2. The model's decision to use the addition function.
3. The result of the function execution.
4. The model's final response.