Model Context Protocol (MCP) servers are a new, standardized way to provide context and tools to your LLMs, and FastMCP makes building MCP servers simple and intuitive. Create tools, expose resources, and define prompts with clean, Pythonic code:

# demo.py

from fastmcp import FastMCP


mcp = FastMCP("Demo 🚀")


@mcp.tool()
def add(a: int, b: int) -> int:
    """Add two numbers"""
    return a + b

That's it! Give Claude access to the server by running:

fastmcp install demo.py

FastMCP handles all the complex protocol details and server management, so you can focus on building great tools. It's designed to be high-level and Pythonic - in most cases, decorating a function is all you need.

• Fast: High-level interface means less code and faster development
• Simple: Build MCP servers with minimal boilerplate
• Pythonic: Feels natural to Python developers
• Complete*: FastMCP aims to provide a full implementation of the core MCP specification

(*emphasis on aims)

🚨 🚧 🏗️ FastMCP is under active development, as is the MCP specification itself. Core features are working but some advanced capabilities are still in progress.

• Installation
• Quickstart
• What is MCP?
• Core Concepts
  • Server
  • Resources
  • Tools
  • Prompts
  • Images
  • Context
• Running Your Server
  • Development Mode (Recommended for Building & Testing)
  • Claude Desktop Integration (For Regular Use)
  • Direct Execution (For Advanced Use Cases)
  • Server Object Names
• Examples
  • Echo Server
  • SQLite Explorer
• Contributing
  • Prerequisites
  • Installation
  • Testing
  • Formatting
  • Opening a Pull Request

We strongly recommend installing FastMCP with uv, as it is required for deploying servers:

uv pip install fastmcp

Note: on macOS, uv may need to be installed with Homebrew (brew install uv) in order to make it available to the Claude Desktop app.

Alternatively, to use the SDK without deploying, you may use pip:

pip install fastmcp

Let's create a simple MCP server that exposes a calculator tool and some data:

# server.py

from fastmcp import FastMCP


# Create an MCP server
mcp = FastMCP("Demo")


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


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

You can install this server in Claude Desktop and interact with it right away by running:

fastmcp install server.py

Alternatively, you can test it with the MCP Inspector:

fastmcp dev server.py

The Model Context Protocol (MCP) lets you build servers that expose data and functionality to LLM applications in a secure, standardized way. Think of it like a web API, but specifically designed for LLM interactions. MCP servers can:

• Expose data through Resources (think of these sort of like GET endpoints; they are used to load information into the LLM's context)
• Provide functionality through Tools (sort of like POST endpoints; they are used to execute code or otherwise produce a side effect)
• Define interaction patterns through Prompts (reusable templates for LLM interactions)
• And more!

There is a low-level Python SDK available for implementing the protocol directly, but FastMCP aims to make that easier by providing a high-level, Pythonic interface.

The FastMCP server is your core interface to the MCP protocol. It handles connection management, protocol compliance, and message routing:

from fastmcp import FastMCP

# Create a named server
mcp = FastMCP("My App")

# Specify dependencies for deployment and development
mcp = FastMCP("My App", dependencies=["pandas", "numpy"])

Resources are how you expose data to LLMs. They're similar to GET endpoints in a REST API - they provide data but shouldn't perform significant computation or have side effects. Some examples:

• File contents
• Database schemas
• API responses
• System information

Resources can be static:

@mcp.resource("config://app")
def get_config() -> str:
    """Static configuration data"""
    return "App configuration here"

Or dynamic with parameters (FastMCP automatically handles these as MCP templates):

@mcp.resource("users://{user_id}/profile")
def get_user_profile(user_id: str) -> str:
    """Dynamic user data"""
    return f"Profile data for user {user_id}"

Tools let LLMs take actions through your server. Unlike resources, tools are expected to perform computation and have side effects. They're similar to POST endpoints in a REST API.

Simple calculation example:

@mcp.tool()
def calculate_bmi(weight_kg: float, height_m: float) -> float:
    """Calculate BMI given weight in kg and height in meters"""
    return weight_kg / (height_m ** 2)

HTTP request example:

import httpx

@mcp.tool()
async def fetch_weather(city: str) -> str:
    """Fetch current weather for a city"""
    async with httpx.AsyncClient() as client:
        response = await client.get(
            f"https://api.weather.com/{city}"
        )
        return response.text

Complex input handling example:

from pydantic import BaseModel, Field
from typing import Annotated

class ShrimpTank(BaseModel):
    class Shrimp(BaseModel):
        name: Annotated[str, Field(max_length=10)]

    shrimp: list[Shrimp]

@mcp.tool()
def name_shrimp(
    tank: ShrimpTank,
    # You can use pydantic Field in function signatures for validation.
    extra_names: Annotated[list[str], Field(max_length=10)],
) -> list[str]:
    """List all shrimp names in the tank"""
    return [shrimp.name for shrimp in tank.shrimp] + extra_names

Prompts are reusable templates that help LLMs interact with your server effectively. They're like "best practices" encoded into your server. A prompt can be as simple as a string:

@mcp.prompt()
def review_code(code: str) -> str:
    return f"Please review this code:\n\n{code}"

Or a more structured sequence of messages:

from fastmcp.prompts.base import UserMessage, AssistantMessage

@mcp.prompt()
def debug_error(error: str) -> list[Message]:
    return [
        UserMessage("I'm seeing this error:"),
        UserMessage(error),
        AssistantMessage("I'll help debug that. What have you tried so far?")
    ]

FastMCP provides an Image class that automatically handles image data in your server:

from fastmcp import FastMCP, Image
from PIL import Image as PILImage

@mcp.tool()
def create_thumbnail(image_path: str) -> Image:
    """Create a thumbnail from an image"""
    img = PILImage.open(image_path)
    img.thumbnail((100, 100))
    
    # FastMCP automatically handles conversion and MIME types
    return Image(data=img.tobytes(), format="png")

@mcp.tool()
def load_image(path: str) -> Image:
    """Load an image from disk"""
    # FastMCP handles reading and format detection
    return Image(path=path)

Images can be used as the result of both tools and resources.

The Context object gives your tools and resources access to MCP capabilities. To use it, add a parameter annotated with fastmcp.Context:

from fastmcp import FastMCP, Context

@mcp.tool()
async def long_task(files: list[str], ctx: Context) -> str:
    """Process multiple files with progress tracking"""
    for i, file in enumerate(files):
        ctx.info(f"Processing {file}")
        await ctx.report_progress(i, len(files))
        
        # Read another resource if needed
        data = await ctx.read_resource(f"file://{file}")
        
    return "Processing complete"

The Context object provides:

• Progress reporting through report_progress()
• Logging via debug(), info(), warning(), and error()
• Resource access through read_resource()
• Request metadata via request_id and client_id

There are three main ways to use your FastMCP server, each suited for different stages of development:

The fastest way to test and debug your server is with the MCP Inspector:

fastmcp dev server.py

This launches a web interface where you can:

• Test your tools and resources interactively
• See detailed logs and error messages
• Monitor server performance
• Set environment variables for testing

During development, you can:

• Add dependencies with --with:

  fastmcp dev server.py --with pandas --with numpy

• Mount your local code for live updates:

  fastmcp dev server.py --with-editable .

Once your server is ready, install it in Claude Desktop to use it with Claude:

fastmcp install server.py

Your server will run in an isolated environment with:

• Automatic installation of dependencies specified in your FastMCP instance:

  mcp = FastMCP("My App", dependencies=["pandas", "numpy"])

• Custom naming via --name:

  fastmcp install server.py --name "My Analytics Server"

• Environment variable management:

  # Set variables individually
  fastmcp install server.py -e API_KEY=abc123 -e DB_URL=postgres://...
  
  # Or load from a .env file
  fastmcp install server.py -f .env

For advanced scenarios like custom deployments or running without Claude, you can execute your server directly:

from fastmcp import FastMCP

mcp = FastMCP("My App")

if __name__ == "__main__":
    mcp.run()

Run it with:

# Using the FastMCP CLI
fastmcp run server.py

# Or with Python/uv directly
python server.py
uv run python server.py

Note: When running directly, you are responsible for ensuring all dependencies are available in your environment. Any dependencies specified on the FastMCP instance are ignored.

Choose this method when you need:

• Custom deployment configurations
• Integration with other services
• Direct control over the server lifecycle

All FastMCP commands will look for a server object called mcp, app, or server in your file. If you have a different object name or multiple servers in one file, use the syntax server.py:my_server:

# Using a standard name
fastmcp run server.py

# Using a custom name
fastmcp run server.py:my_custom_server

Here are a few examples of FastMCP servers. For more, see the examples/ directory.

A simple server demonstrating resources, tools, and prompts:

from fastmcp import FastMCP

mcp = FastMCP("Echo")

@mcp.resource("echo://{message}")
def echo_resource(message: str) -> str:
    """Echo a message as a resource"""
    return f"Resource echo: {message}"

@mcp.tool()
def echo_tool(message: str) -> str:
    """Echo a message as a tool"""
    return f"Tool echo: {message}"

@mcp.prompt()
def echo_prompt(message: str) -> str:
    """Create an echo prompt"""
    return f"Please process this message: {message}"

A more complex example showing database integration:

from fastmcp import FastMCP
import sqlite3

mcp = FastMCP("SQLite Explorer")

@mcp.resource("schema://main")
def get_schema() -> str:
    """Provide the database schema as a resource"""
    conn = sqlite3.connect("database.db")
    schema = conn.execute(
        "SELECT sql FROM sqlite_master WHERE type='table'"
    ).fetchall()
    return "\n".join(sql[0] for sql in schema if sql[0])

@mcp.tool()
def query_data(sql: str) -> str:
    """Execute SQL queries safely"""
    conn = sqlite3.connect("database.db")
    try:
        result = conn.execute(sql).fetchall()
        return "\n".join(str(row) for row in result)
    except Exception as e:
        return f"Error: {str(e)}"

@mcp.prompt()
def analyze_table(table: str) -> str:
    """Create a prompt template for analyzing tables"""
    return f"""Please analyze this database table:
Table: {table}
Schema: 
{get_schema()}

What insights can you provide about the structure and relationships?"""

git clone https://github.com/jlowin/fastmcp.git
cd fastmcp
uv sync --frozen --extra dev

uv sync --frozen --extra tests

pytest -vv

pre-commit install

pre-commit run --all-files

git checkout -b my-branch

git add . && git commit -m "My changes"

git push origin my-branch