FastMCP — Python Framework for MCP Servers

Main

• 先从 README 的最小 server 示例跑通，再把每个工具拆成可测试、带清晰 docstring 的 Python 函数。

• 把工具的 schema/返回值当作 API：保持契约稳定，对破坏性变更做版本化管理。

• 尽早补齐生产化要素（超时、限流、鉴权），并准备一个 staging 环境复核不同 MCP 客户端的兼容性。

• 优先按文档里的 server/client/app 模式扩展，不要手写一堆协议胶水。

README (excerpt)

FastMCP 🚀

Move fast and make things.

Made with 💙 by Prefect

The Model Context Protocol (MCP) connects LLMs to tools and data. FastMCP gives you everything you need to go from prototype to production:

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()

Why FastMCP

Building an effective MCP application is harder than it looks. FastMCP handles all of it. Declare a tool with a Python function, and the schema, validation, and documentation are generated automatically. Connect to a server with a URL, and transport negotiation, authentication, and protocol lifecycle are managed for you. You focus on your logic, and the MCP part just works: with FastMCP, best practices are built in.

Source-backed notes

• README 给出最小 server 示例：FastMCP("Demo")、@mcp.tool、以及 mcp.run() 启动。
• README 推荐用 uv pip install fastmcp 安装，并链接到独立文档站点。
• README 描述 FastMCP 用于抽象 MCP 的 schema/生命周期细节，减少手写协议逻辑。

FAQ

• FastMCP 是 server 还是 client？：README 同时提到 server/client 模式；建议先从 server 落地开始。
• 必须用 uv 吗？：README 推荐 uv；但作为 Python 依赖也可以按团队习惯使用 pip/虚拟环境。
• 怎么保证工具行为可预测？：把 schema 当作契约，加测试，并对破坏性改动做版本化发布。