MCP Server Template (python)

🚀 MCP Server Template (Python)

A ready-to-use template for quickly building a Model Context Protocol (MCP) server. This template helps you rapidly create a server that can register and expose tools and prompts for AI models.

🚀 Quick Start

Prerequisites

• Python 3.10 or higher

Three-Step Quick Setup

1️⃣ Install the Package

# Clone the repository
git clone https://github.com/nisarg38/mcp-server-template-python.git my-mcp-server
cd my-mcp-server

# Install in development mode
pip install -e ".[dev]"

2️⃣ Run Your Server

# Run with Python
python -m src.main

# Or use the convenient CLI
mcp-server-template

3️⃣ Your Server is Now Up and Running!

Access your MCP server:

• 🌐 HTTP: http://localhost:8080
• 💻 Or use stdio transport: mcp-server-template --transport stdio

You'll see log output confirming that the server is running successfully.

✨ Features

🎮 Command Line Options

Customize your server's behavior with these command-line options:

# Change the port (default: 8080)
mcp-server-template --port 9000

# Enable debug mode for more detailed logs
mcp-server-template --debug

# Use stdio transport instead of HTTP
mcp-server-template --transport stdio

# Set the log level (options: debug, info, warning, error)
mcp-server-template --log-level debug

🛠️ Creating Your Own Tools and Prompts

Adding a Tool

Tools are functions that AI models can call. To add a new tool:

1. Edit src/main.py
2. Add a new function using the @mcp.tool() decorator:

@mcp.tool()
def your_tool_name(param1: str, param2: int) -> Dict[str, Any]:
    """
    Example: Add a description of the tool here.

    Args:
        param1: Describe parameter 1.
        param2: Describe parameter 2.

    Returns:
        A description of the return value.
    """

Adding a Prompt

Prompts are strings that models can use to generate responses. To add a new prompt:

1. Edit src/main.py
2. Add a new function using the @mcp.prompt() decorator:

@mcp.prompt()
def your_prompt_name() -> str:
    """
    Example: Add a description of the prompt here.
    """

💻 Usage Examples

Basic Usage

@mcp.tool()
def add_numbers(param1: float, param2: float) -> float:
    """
    Args:
        param1: The first number.
        param2: The second number.

    Returns:
        The sum of the two numbers.
    """
    return param1 + param2

@mcp.prompt()
def math_prompt() -> str:
    """
    Returns:
        "Please calculate the sum of these two numbers."
    """

📚 Documentation

📦 Project Structure

your_project/
├── src/
│   ├── main.py         # Main file for the MCP server
│   └── prompts.py      # Where prompts are defined
└── requirements.txt   # List of necessary dependencies

💻 Deployment Options

Web-Based Interface

You can create a simple web-based interface to interact with the MCP server using FastAPI or Flask.

Packaging and Deployment

Package your MCP server as a standalone executable using setuptools and then deploy it to a production environment.

🔧 Development Guide

Install Dependencies

List all necessary libraries in requirements.txt and run:

pip install -r requirements.txt

Testing and Debugging

Write unit tests to verify that your tools and prompts work as expected, and then use logging and debugging information for troubleshooting.

🤝 Need Help?

• Ask a question: Open an issue on GitHub.
• Join the community: Join our Discord community (coming soon).

Made with ❤️ for the AI developer community.