FastMCP SDKs
Create powerful Model Context Protocol servers that expose data and functionality to AI applications. FastMCP provides a "USB-C port for AI" with standardized resource and tool interactions.
Developer Tools for MCP
FastMCP eliminates boilerplate code and provides intuitive APIs for building MCP servers. Create tools, define resources, and generate prompts with built-in authentication and validation.
Pythonic & TypeScript-First
High-level, intuitive interfaces designed for developer productivity with minimal boilerplate.
Production Ready
Built-in authentication, session management, and support for multiple transport protocols (STDIO, HTTP, SSE).
Comprehensive Ecosystem
Full ecosystem with OpenAPI generation, server composition, testing tools, and CLI support.
FastMCP Python
Python framework for building MCP servers with decorators and async support
FastMCP TypeScript
TypeScript framework built on the official SDK with zero boilerplate
Getting Started
Both SDKs provide the same powerful features with language-specific optimizations. Get started in minutes with these simple examples.
Python Quick Start
# Install with uv (recommended)
uv pip install fastmcp
# Create your first server
from fastmcp import FastMCP
mcp = FastMCP("Demo 🚀")
@mcp.tool
def add(a: int, b: int) -> int:
"""Add two numbers"""
return a + b
if __name__ == "__main__":
mcp.run()
TypeScript Quick Start
// Install the SDK
npm install fastmcp
// Create your first server
import { FastMCP } from "fastmcp"
import { z } from "zod"
const server = new FastMCP({
name: "My Server",
version: "1.0.0"
})
server.addTool({
name: "add",
description: "Add two numbers",
parameters: z.object({
a: z.number(), b: z.number()
}),
execute: async (args) => String(args.a + args.b)
})
server.start({ transportType: "stdio" })
Join the Community
Get help, share your projects, and contribute to the FastMCP ecosystem. We're here to support you on your MCP journey.