🌦️ Weather MCP (Model Context Protocol) Demo

🌦️ Weather MCP (Model Context Protocol) Demo

This project demonstrates the implementation of a Model Context Protocol (MCP) system for weather information, consisting of a client using OpenAI's agents and a dedicated MCP server. The system allows users to query weather information through a natural language interface powered by GPT-4.

---

🌟 Overview

The project is split into two main components:

• Weather MCP Server: A Python-based server that provides weather information.
• Weather OpenAI Client: A client that uses OpenAI's GPT-4 model to process natural language requests and interact with the MCP server.

---

🚀 Components

🌐 Weather MCP Server

📁 Located in: /server_weather/

The server component implements a Model Context Protocol server that:

• Exposes a weather information endpoint via SSE (Server-Sent Events).
• Provides a fetch-weather tool that can be discovered by MCP clients.
• Runs on http://localhost:4000/sse.

🤖 Weather OpenAI Client

📁 Located in: /client_weather_openai/

The client component features:

• Integration with OpenAI's GPT-4 model.
• Automatic tool discovery from the MCP server.
• Natural language processing for weather queries.
• Interactive command-line interface for user interactions.

---

📋 Prerequisites

• Python 3.x
• OpenAI API Key (Get yours from OpenAI Platform)
• Environment Variables properly configured (.env file)

---

🔧 Setup

1. Clone this repository

git clone [email protected]:sturlese/client_server_mcp.git
cd client_server_mcp

2. Set up environment variables

• Create a .env file in the root of each component (/server_weather/ and /client_weather_openai/).
• Add your OpenAI API key:

# Example of .env file
OPENAI_API_KEY='your-api-key-here'

3. Install dependencies

Server component:

cd server_weather/
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

Client component:

cd ../client_weather_openai/
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

---

🎮 Usage

Start the MCP Server

cd server_weather/
source venv/bin/activate
python server_weather.py

Start the OpenAI Client

cd client_weather_openai/
source venv/bin/activate
python weather_agent_openai.py

Example Usage

What's the weather in New York today?

---

🏗️ Architecture

The system uses the Model Context Protocol (MCP) to enable seamless communication between the AI agent and the weather service:

User ---> OpenAI Agent (Client) ---> MCP Server ---> Weather Data

---

📝 License

This project is open-source and available under the MIT License.

---

🤝 Contributing

Contributions, issues, and feature requests are welcome!

1. Fork the project.
2. Create a new branch (git checkout -b feature/your-feature).
3. Make your changes.
4. Commit your changes (git commit -m 'Add your feature').
5. Push to your branch (git push origin feature/your-feature).
6. Open a pull request.

---