Skip to main content
Learn MCP through games

Master MCP by Playing

Interactive challenges that teach you how AI agents communicate with tools. Play games, solve puzzles, learn the protocol.

Start Playing
9+
Challenges
Free
Forever
Open
Source

Featured Challenges

Start with these popular games to learn MCP fundamentals

Try it Now

No signup needed. Just play.

Want an AI agent to play for you?

Browse Challenges

How It Works

MCP lets AI agents use tools through a simple, standardized protocol

Step 1

Pick a Challenge

Choose from chess, puzzles, drawing games, and more. Each game teaches different MCP concepts.

Step 2

Connect via MCP

Use tools like make_move, draw_pixel, or get_state. Your AI client communicates through the protocol.

Step 3

Learn by Playing

Watch how tool calls translate to actions. Understand MCP patterns through real interactions.

mcp-server.ts
// Define an MCP tool for your AI agent
server.tool(
  "chess.make_move",
  "Make a chess move on the board",
  {
    from: z.string().describe("Square to move from (e.g., 'e2')"),
    to: z.string().describe("Square to move to (e.g., 'e4')")
  },
  async ({ from, to }) => {
    const result = await game.move({ from, to });
    return {
      success: result.valid,
      board: game.ascii(),
      status: game.isGameOver() ? "Game Over" : "In Progress"
    };
  }
);
claude_desktop_config.json
// Paste into Claude Desktop settings to start playing
{
  "mcpServers": {
    "mcpchallenge-chess": {
      "url": "https://mcp.mcpchallenge.org/chess/sse"
    }
  }
}

Add this to your Claude Desktop config, then ask Claude to play chess.

9+
Thử thách
50+
Achievements
Khả năng
$0
Chi phí

Ready to start learning?

Jump into your first challenge and discover how MCP enables AI agents to interact with the world.

Browse Challenges