turn-based-game-mcp: Instructions file for GitHub Copilot

MCP Server Development

Purpose

Patterns for the Model Context Protocol server. Covers tool definitions, HTTP integration with the web API, AI algorithms, and error handling.

Server Architecture

Tool Definition Pattern

Each game operation should be exposed as a separate MCP tool:

server.setRequestHandler(ListToolsRequestSchema, async () => ({
  tools: [
    {
      name: "create_tic_tac_toe_game",
      description: "Create a new Tic-Tac-Toe game with optional custom game ID",
      inputSchema: {
        type: "object",
        properties: {
          gameId: { type: "string", description: "Optional custom game ID" },
          playerName: { type: "string", default: "Player" },
          aiDifficulty: { 
            type: "string", 
            enum: ["easy", "medium", "hard"],
            default: "medium"
          }
        }
      }
    },
    {
      name: "play_tic_tac_toe",
      description: "Make an AI move in Tic-Tac-Toe game",
      inputSchema: {
        type: "object",
        properties: {
          gameId: { type: "string", description: "The ID of the game to play" }
        },
        required: ["gameId"]
      }
    }
  ]
}))

HTTP Integration Pattern

Use HTTP calls to communicate with the web API, not direct database access:

// Helper functions for API communication
async function httpGet(url: string): Promise<any> {
  const response = await fetch(url)
  if (!response.ok) {
    throw new Error(`HTTP ${response.status}: ${response.statusText}`)
  }
  return response.json()
}

async function httpPost(url: string, data: any): Promise<any> {
  const response = await fetch(url, {
    method: 'POST',
    headers: { 'Content-Type': 'application/json' },
    body: JSON.stringify(data)
  })
  if (!response.ok) {
    throw new Error(`HTTP ${response.status}: ${response.statusText}`)
  }
  return response.json()
}

Game Integration Workflow

Game State Retrieval

Always fetch the latest game state before making AI moves: