FastMCP
by jlowin
FastMCP is a Python library that simplifies building Model Context Protocol (MCP) servers and clients. It provides a high-level, Pythonic interface for creating tools, exposing resources, defining prompts, and connecting components for LLM applications.
Last updated: N/A
What is FastMCP?
FastMCP is a Python library designed to facilitate the creation of MCP (Model Context Protocol) servers and clients. It abstracts away the complexities of the MCP protocol, allowing developers to focus on building tools and applications that interact with LLMs in a standardized and secure manner.
How to use FastMCP?
To use FastMCP, you install the library using uv pip install fastmcp. You then define your server using the FastMCP class and decorate your functions with @mcp.tool(), @mcp.resource(), and @mcp.prompt() to expose functionality to LLMs. You can run the server using the fastmcp run command or integrate it into Claude Desktop using fastmcp install.
Key features of FastMCP
Simplified server creation with decorators
Client library for interacting with MCP servers
Support for tools, resources, and prompts
Advanced features like proxy servers and server composition
Automatic generation of servers from OpenAPI specs or FastAPI objects
Use cases of FastMCP
Exposing data and functionality to LLM applications
Building custom tools for LLMs
Creating reusable interaction patterns for LLMs
Integrating existing web APIs with LLMs
FAQ from FastMCP
What is MCP?
What is MCP?
The Model Context Protocol (MCP) lets you build servers that expose data and functionality to LLM applications in a secure, standardized way.
Why use FastMCP?
Why use FastMCP?
FastMCP handles the complex protocol details and server management, so you can focus on building great tools. It's designed to be high-level and Pythonic.
How do I install FastMCP?
How do I install FastMCP?
We strongly recommend installing FastMCP with uv: uv pip install fastmcp
How do I define a tool?
How do I define a tool?
Decorate a Python function with @mcp.tool(). FastMCP automatically generates the necessary MCP schema based on type hints and docstrings.
How do I define a resource?
How do I define a resource?
Decorate a Python function with @mcp.resource("your://uri"). Use curly braces {} in the URI to define dynamic resources.