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.
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-itscope 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
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
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.
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.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
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.
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.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
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.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.