Hello MCP Server Python

🚀 MCP Server Documentation

This project is a server-side tool implemented using the Model Context Protocol (MCP), aiming to provide functional support for large language models (LLMs) such as Claude AI. Currently, the server offers two main services: BMI calculation and weather information query.

🚀 Quick Start

This MCP server is designed to support large language models by offering BMI calculation and weather information query services.

✨ Features

• BMI Calculation: The calculate_bmi tool accepts height (cm) and weight (kg) as inputs and outputs the BMI index.
• Weather Information Query: The fetch_weather tool queries weather information based on the city name.

📦 Installation

System Requirements

• Operating System: It is recommended to use Linux or macOS. Windows can also run, but some functions may be limited.
• Python Version: Python 3.9 or higher.
• Development Environment: Basic knowledge of Python and command-line operations is required.

Installation Steps

1. Clone the code repository:

   git clone https://github.com/yourusername/hello-mcp-server-python.git
   cd hello-mcp-server-python
   

2. Install dependencies:

   pip install httpx "mcp[cli]"
   

3. Start the server (optional, run via MCP or command line):

   python server.py
   # Or use MCP CLI
   mcp run server.py
   

📚 Documentation

Server Configuration

MCP Configuration File (mcp.json)

Create or modify the mcp.json file in the project root directory and add the following content to configure the server:

{
  "mcpServers": {
      "mcp-server-python": {
            "command": "mcp",
            "args": [
                "run",
                "/path/to/server.py"
                ]
      }
  }
}

Command-Line Execution

Start the server by directly executing the following command in the terminal:

python server.py

Environment Variable Configuration

API Key Setup

Add the following code to server.py to configure the weather API key:

API_KEY = "your_api_key_here"  # Replace with the actual API key

Project Structure

hello-mcp-server-python/
├── server.py       # Main program file of the MCP server
└── requirements.txt  # Dependency management file (optional)

Usage Instructions

Connect via MCP

1. Open the Cursor IDE and load the project.
2. Ensure that the mcp.json file exists and is correctly configured.
3. Run the server in Cursor, and the MCP tools will be automatically recognized and available.

Use via Command Line

1. Install Claude for Desktop.
2. Start the server in the terminal:

   python server.py
   

3. Access the MCP server in the Claude interface for interaction.

Frequently Asked Questions

• How to obtain the weather API key?
  • You can contact the relevant weather service provider to apply for a free trial key.
• What if the server fails to start?
  • Check the network connection and whether the dependencies are installed correctly, and view the error information output in the terminal.

💻 Usage Examples

Basic Usage

• BMI Calculation:

  calculate_bmi --height 170 --weight 65
  

  The output is the BMI value.
• Weather Query:

  fetch_weather --city "Beijing"
  

  It returns the weather details of the specified city.

📄 License

This project is licensed under the MIT License. Please refer to the LICENSE file for detailed information.

📚 References

• Model Context Protocol Documentation
• MCP Python SDK Documentation
• httpx Official Documentation
• Claude AI Development Documentation