MCP (Model Context Protocol) Chat Application

This is a chat application that implements the Model Context Protocol (MCP) to enable communication between a client and server with AI model integration. The application uses Google's Gemini AI model and provides a command-line interface for interaction.

Features

• Real-time communication between client and server using MCP
• Integration with Google's Gemini AI model
• Command-line interface for user interaction
• Tool support for extending functionality
• Session management with unique IDs
• CORS support for cross-origin requests

Prerequisites

• Node.js (v14 or higher)
• npm (Node Package Manager)
• Google Gemini API key

Project Structure

.
├── client/             # Client-side application
│   ├── index.js       # Main client implementation
│   └── package.json   # Client dependencies
├── server/            # Server-side application
│   ├── index.js      # Main server implementation
│   └── package.json  # Server dependencies
└── package.json      # Root project dependencies

Setup

1. Clone the repository:

   git clone <repository-url>
   cd mcp-trying
   

2. Install dependencies:

   # Install root dependencies
   npm install
   
   # Install server dependencies
   cd server
   npm install
   
   # Install client dependencies
   cd ../client
   npm install
   

3. Create a .env file in the client directory with your Google Gemini API key:

   GEMINI_API_KEY=your_api_key_here
   

Running the Application

1. Start the server:

   cd server
   node index.js
   

   The server will start on port 3000 by default.

2. In a new terminal, start the client:

   cd client
   node index.js
   

Usage

1. Once both the server and client are running, you can start chatting through the command-line interface.
2. Type your message and press Enter to send it to the AI model.
3. The AI will respond and can use available tools if needed.
4. The conversation history is maintained during the session.

Available Tools

The server currently implements the following tool:

• addTwoNumbers: Adds two numbers together and returns the sum

Technical Details

• The application uses the Model Context Protocol (MCP) for communication
• Server implements Express.js for handling HTTP requests
• Client uses readline for command-line interaction
• Google's Gemini AI model is used for generating responses
• Session management is handled using UUIDs
• CORS is enabled for cross-origin requests

Error Handling

The application includes error handling for:

• Connection issues
• Invalid session IDs
• Server errors
• Chat loop errors

Contributing

Feel free to submit issues and enhancement requests!

License

This project is licensed under the MIT License - see the LICENSE file for details.