Ujjwal3115/MCP_Server_Model_Twitter

MCP Server Example

This project demonstrates a simple client-server application using the Model Context Protocol (MCP) with Google Gemini AI and Twitter API integration.

Features

• Server: Exposes tools via MCP, including:
  • addTwoNumbers: Adds two numbers.
  • createPost: Posts a status to X (formerly Twitter).
• Client: Connects to the MCP server, lists available tools, and interacts with Gemini AI to call tools.

Prerequisites

• Node.js v18 or higher
• npm
• Twitter Developer credentials (for posting to X)
• Google Gemini API key

Setup

1. Clone the repository

   git clone <your-repo-url>
   cd <repo-folder>
   

2. Install dependencies

   cd server
   npm install
   cd ../client
   npm install
   

3. Environment Variables Create a .env file in both server/ and client/ directories with the following variables: cles server/.env

   TWITTER_API_KEY=your_twitter_api_key
   TWITTER_API_SECRET=your_twitter_api_secret
   TWITTER_ACCESS_TOKEN=your_twitter_access_token
   TWITTER_ACCESS_TOKEN_SECRET=your_twitter_access_token_secret
   

   client/.env

   GEMINI_API_KEY=your_gemini_api_key
   

Running the Project

1. Start the server

   cd server
   node indexs.js
   

   The server will run at http://localhost:3001.

2. Start the client In a new terminal:

   cd client
   node indexc.js
   

3. Usage

   • The client will connect to the server and prompt you for input.
   • You can ask questions or trigger tool calls (e.g., add two numbers, post a status to X).

Project Structure

client/
  indexc.js
  package.json
server/
  indexs.js
  mcp.tool.js
  package.json

Notes

• Ensure your Twitter and Gemini API credentials are valid.
• The server must be running before starting the client.
• For production, secure your environment variables and API keys.

License

MIT