Provides a real-time multiplayer Tic-tac-toe game that allows AI agents to connect and play using natural language via a Model Context Protocol server.
This project offers a delightful Tic-tac-toe game with comprehensive real-time multiplayer capabilities, supporting multiple sessions and custom symbols for an engaging user experience. Its unique selling point is the integrated Model Context Protocol (MCP) Server, which empowers AI agents like Copilot or Cursor to interact with and play the game using natural language commands. The system is architected as a Turborepo, separating frontend, backend, and MCP server components, ensuring modularity and scalability for both human and AI-driven gameplay.
Key Features
01Interactive sound effects and visual winning indicators
02Model Context Protocol (MCP) Server for AI agent integration
03Real-time multiplayer gameplay with Socket.IO
04Supports multiple game sessions and custom symbols
05AI agents can play using natural language
060 GitHub stars
Use Cases
01Demonstrate MCP integration with real-time web applications
02Play Tic-tac-toe against an AI agent
03Develop and test natural language-driven AI agents for game interaction
