MCP-IT Fastify

MCP-IT Fastify

A Fastify plugin that automatically converts API routes into Model Context Protocol tools, enabling AI assistants to directly interact with your API.

Category
Visit Server

README

@mcp-it/fastify

🤖 Automatically generate MCP tools from your Fastify API routes.

A Fastify plugin (@mcp-it/fastify) for the Model Context Protocol (MCP) that allows you to expose your Fastify routes as MCP tools. This enables AI assistants to interact with your API directly through the MCP protocol.

Overview

This plugin automatically discovers your Fastify routes and exposes them as tools consumable by MCP clients like Cursor or Claude. It leverages Fastify's schema system to generate complete tool definitions.

[!NOTE] While this package focuses specifically on Fastify, the @mcp-it scope may host adapters for other Node.js frameworks (like Express, NestJS, etc.) in the future.

Installation

npm install @mcp-it/fastify
# or
yarn add @mcp-it/fastify
# or
pnpm add @mcp-it/fastify

Usage

import Fastify from "fastify";
import mcpPlugin from "@mcp-it/fastify";

const fastify = Fastify();

// Register the MCP plugin
await fastify.register(mcpPlugin, {
  name: "My API",
  description: "My API with MCP support",
});

// Define your routes with schemas and operation IDs
fastify.get(
  "/users/:id",
  {
    schema: {
      operationId: "get_user", // Used as the tool name
      summary: "Get user by ID",
      description: "Returns a user by their ID",
      params: {
        type: "object",
        required: ["id"],
        properties: {
          id: { type: "number", description: "User ID" },
        },
      },
      response: {
        200: {
          description: "Successful response",
          type: "object",
          properties: {
            id: { type: "number" },
            name: { type: "string" },
            email: { type: "string" },
          },
        },
      },
    },
    // Add MCP-specific config if needed, e.g.:
    // config: {
    //   mcp: { hidden: true }
    // }
  },
  async (request) => {
    // Implementation...
    const userId = (request.params as any).id;
    // ... fetch user
    return { id: userId, name: "Example User", email: "user@example.com" };
  }
);

await fastify.listen({ port: 3000 });

console.log("MCP SSE server running at http://localhost:3000/mcp/sse");

Features

  • Automatic Route Discovery: Utilizes Fastify hooks to identify all registered routes.
  • Schema Utilization: Employs Fastify route schemas for comprehensive tool generation.
  • Per-Route Configurations: Allows customization on a per-route basis.
  • Multiple Transports: Supports both Server-Sent Events and Streamable HTTP transports.
  • Debug Endpoint: An optional endpoint to view generated tools, independent of transport.

Configuration Options

Option Type Default Description
name string "Fastify MCP" Name for the MCP server displayed to the client.
description string "MCP server for Fastify" Description for the MCP server.
transportType string "sse" Transport protocol to use: "sse" or "streamableHttp".
describeFullSchema boolean false Include detailed input/output schemas and examples in descriptions.
skipHeadRoutes boolean true Exclude HEAD routes from the generated MCP tools.
skipOptionsRoutes boolean true Exclude OPTIONS routes from the generated MCP tools.
mountPath string "/mcp" Base path prefix where MCP SSE and message endpoints are mounted.
filter Function undefined Custom function (route: Route) => boolean for filtering.
addDebugEndpoint boolean false Add a GET /<mountPath>/tools endpoint listing generated tools.

Route Configuration (config.mcp)

You can add specific MCP configurations directly within a route's config object. The following options are available:

Option Type Default Description
hidden boolean false Hide this route from the MCP Server
name string operationId or method_url Override the default tool name
description string Route's schema description Override the tool description

Example usage:

fastify.get(
  "/some-route",
  {
    config: {
      mcp: {
        name: "custom_tool_name", // Override the default tool name
        description: "Custom description for this tool", // Override the default description
      },
    },
    // ... other route options
  },
  async (request, reply) => {
    /* ... */
  }
);

Accessing the MCP Server Instance

This plugin decorates the Fastify instance with the underlying MCP Server instance from @modelcontextprotocol/sdk. You can access it via fastify.mcpServer after the plugin has been registered.

This might be useful for advanced scenarios, such as:

  • Directly interacting with the MCP server's lifecycle or methods.
  • Adding custom request handlers or capabilities beyond basic tool exposure.
import Fastify from "fastify";
import mcpPlugin from "@mcp-it/fastify";
import type { Server } from "@modelcontextprotocol/sdk/server/index.js";

const fastify = Fastify();

await fastify.register(mcpPlugin, {
  /* options */
});

// Now you can access the MCP server instance
console.log("MCP Server:", fastify.mcpServer);

// Example: Add a custom handler (use with caution)
// fastify.mcpServer.setRequestHandler(...);

// ... rest of your application setup

await fastify.listen({ port: 3000 });

Examples

See the examples directory for complete working examples demonstrating various features.

Client Configuration

Cursor

In Cursor settings (Settings -> MCP), add a new SSE connection with the URL:

http://localhost:3000/mcp/sse

(Replace localhost:3000 with your server's address and mcp with your mountPath if customized).

Claude Desktop

Use mcp-proxy to bridge between Claude Desktop (which expects stdio) and the SSE endpoint.

Add the following to your Claude Desktop MCP config file (claude_desktop_config.json):

{
  "mcpServers": {
    "my-fastify-api": {
      "command": "npx",
      "args": ["mcp-remote", "http://localhost:3000/mcp/sse"]
    }
  }
}

(Replace localhost:3000 with your server's address and mcp with your mountPath if customized).

Alternatively, you can use the Streamable HTTP endpoint if your client supports it (Cursor might require configuration or proxying for non-SSE endpoints).

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured