fastify

This MCP server provides a high-performance Model Context Protocol implementation built with Fastify and TypeScript. It exposes an MCP-compliant API with endpoints for health checks and metrics, and automatically discovers and registers tools, resources, and prompts from your project folders. Access to the MCP endpoints is secured via a Bearer token configured in MCP_SERVER_TOKEN, ensuring secure connections for AI agents and LLM integrations. Use the included MCP transport (WebSocket/HTTP) to interact with tools, resources, and prompts in a structured, type-safe manner through the @modelcontextprotocol/sdk.

Once running, you can interact with the MCP server through its standard endpoints and the MCP client tooling. The auto-discovery mechanism scans the tools, resources, and prompts directories to register capabilities without manual wiring, making it easy to extend capabilities as your project grows. The server also includes production-ready features such as metrics, health checks, and session management to support long-running deployments in Kubernetes or other orchestration environments.

Prerequisites:

• Node.js 20.10.0+ (recommended 24.x LTS)
• npm or yarn
• Git

Installation steps:

1. Clone the repository:

git clone https://github.com/your-username/neddl-fastify-mcp-server.git
cd neddl-fastify-mcp-server

2. Install dependencies:

npm install

3. Build the project (TypeScript to JavaScript):

npm run build

4. Run the server in development or production mode (see environment variables in configuration):

# Development (with hot reload if configured)
npm run dev

# Production (standalone MCP server)
npm start

5. Compose the runtime environment variables (example):

cp .env.example .env

Edit .env with your configuration, including MCP_SERVER_PORT, MCP_SERVER_HOST, and MCP_SERVER_TOKEN.

Tips and notes:

• Ensure MCP_SERVER_TOKEN is kept secret and rotated regularly in production.
• The MCP auto-discovery relies on the presence of Tools, Resources, and Prompts in their respective folders and updates when you modify index.ts in those directories.
• If running in Kubernetes, expose port 9080 (as per .env.example) and configure readiness/liveness probes to monitor /health and /metrics.
• For local development, you can use the dev script (npm run dev) to speed up iteration; for production, use npm start and a process manager like PM2 or a Kubernetes deployment.
• If you encounter TypeScript compilation issues, verify your Node.js version matches the supported range and that dependencies are installed correctly.