mcp-server-playbook-2025

The MCP Python SDK provides tools to build MCP servers and clients, facilitating seamless integration between Large Language Models (LLMs) and external data sources or tools. This SDK adheres to the full MCP specification, ensuring compatibility and standardization. It allows developers to create servers that can host various tools and resources, enabling AI models to access and utilize these external functionalities effectively. The SDK is designed to be user-friendly, with features that support quick setup and deployment, making it ideal for both development and production environments.

Features

• Seamless integration with LLMs and external tools
• Adherence to full MCP specification for compatibility
• User-friendly setup and deployment process
• Support for dynamic resources and tools
• Facilitates secure and scalable server deployment

MCP Tools

• add: Adds two numbers.

MCP Resources

• {'name': 'greeting://{name}', 'description': 'Generates a personalized greeting.'}

Usage with Different Platforms

mcp

bash
uv add "mcp[cli]"

pip

bash
pip install mcp

server

python
from mcp.server.fastmcp import FastMCP

# Initialize the MCP server
mcp = FastMCP("Demo Server")

# Define an addition tool
@mcp.tool()
def add(a: int, b: int) -> int:
    """Adds two numbers."""
    return a + b

# Define a dynamic greeting resource
@mcp.resource("greeting://{name}")
def get_greeting(name: str) -> str:
    """Generates a personalized greeting."""
    return f"Hello, {name}!"

# Run the server
if __name__ == "__main__":
    mcp.run()

testing

bash
mcp dev server.py

client_integration

bash
mcp install server.py