MCP Server

This project is based on the Web Dev Simplified tutorial: The Ultimate MCP Crash Course - Build From Scratch.

MCP Server is a Node.js/TypeScript project implementing a server compatible with the Model Context Protocol (MCP), designed for AI and automation applications. The project includes both the server logic and a sample client CLI.

Main Features

• MCP Server: Implements the MCP protocol for handling AI requests.
• Sample Client CLI: A command-line interface to test and interact with the server, built using @inquirer/prompts for interactive prompts and input.
• Gemini API Support: Integration with Google Gemini via @ai-sdk/google.
• Data Validation: Uses Zod for input validation.
• Configuration Management: Supports environment variables via dotenv.

Project Structure

├── src/
│   ├── server.ts        # MCP server entry point
│   ├── client.ts        # Sample client
│   └── data/
│       └── users.json   # Sample user data
├── build/               # TypeScript build output
├── package.json         # Dependencies and scripts
├── tsconfig.json        # TypeScript configuration
├── .env.example         # Example environment configuration

Prerequisites

• Node.js >= 18
• npm >= 9

Installation

1. Clone the repository:

   git clone <repo-url>
   cd mcp-server

2. Install dependencies:

   npm install

3. Configure environment variables:
   • Copy .env.example to .env and add your Gemini API key.

Main Scripts

• Start server in development:

  npm run server:dev

• Build the server:

  npm run server:build

• Run sample client:

  npm run client:dev

• MCP Inspector (debug):

  npm run server:inspect

Main Dependencies

• @modelcontextprotocol/sdk: MCP SDK for protocol implementation
• @ai-sdk/google: Gemini integration for AI requests
• zod: Data validation for input and output
• dotenv: Environment variable management
• tsx: TypeScript execution
• @inquirer/prompts: Used to build the interactive CLI client, providing user-friendly command-line prompts and input handling

Environment Configuration

Create a .env file based on .env.example:

GEMINI_API_KEY=your_gemini_api_key

Get your key from Google Gemini API.

Additional Notes

• The sample client is a CLI application located in src/client.ts, using @inquirer/prompts for interactive command-line functionality.
• Sample user data is in src/data/users.json.
• Server and client are written in TypeScript and compiled to build/.
• To customize logic, edit src/server.ts and src/client.ts.
• To stop the application, use the key combination Ctrl+Cin the terminal.

References

• Model Context Protocol
• Google Gemini API