Say MCP Server

Instructions

Implementation Reference

• src/index.ts:114-131 (handler)

  Handler for the list_voices tool: executes 'say -v "?"' to list macOS TTS voices and returns the stdout as text content, with error handling.

  case 'list_voices': {
    try {
      const { stdout } = await execAsync('say -v "?"');
      return {
        content: [
          {
            type: 'text',
            text: stdout,
          },
        ],
      };
    } catch (error: any) {
      throw new McpError(
        ErrorCode.InternalError,
        `Failed to list voices: ${error.message}`
      );
    }
  }

• src/index.ts:75-82 (registration)

  Registration of the list_voices tool in the ListToolsRequestSchema response, including name, description, and input schema (no required params).

  {
    name: 'list_voices',
    description: 'List available text-to-speech voices',
    inputSchema: {
      type: 'object',
      properties: {},
    },
  },

• src/index.ts:78-81 (schema)

  Input schema for list_voices tool: empty object schema (no parameters).

  inputSchema: {
    type: 'object',
    properties: {},
  },