Attendee MCP Server

Instructions

Implementation Reference

• src/index.ts:527-561 (handler)

  The main handler function that validates inputs, constructs the speech data, makes the API POST request to /api/v1/bots/{bot_id}/speech, and returns a success message.

  private async makeBotSpeak(args: Record<string, unknown>) {
    const bot_id = args.bot_id as string;
    const text = args.text as string;
    const voice_language_code = (args.voice_language_code as string) || "en-US";
    const voice_name = (args.voice_name as string) || "en-US-Casual-K";
    
    if (!bot_id || typeof bot_id !== 'string') {
      throw new Error("Missing or invalid required parameter: bot_id");
    }
    
    if (!text || typeof text !== 'string') {
      throw new Error("Missing or invalid required parameter: text");
    }
    
    const speechData = {
      text,
      text_to_speech_settings: {
        google: {
          voice_language_code,
          voice_name
        }
      }
    };
  
    await this.makeApiRequest(`/api/v1/bots/${bot_id}/speech`, "POST", speechData);
  
    return {
      content: [
        {
          type: "text",
          text: `✅ Bot ${bot_id} will speak: "${text}"\n\n🔊 Voice: ${voice_name} (${voice_language_code})\n💡 The bot should now be speaking in the meeting!`,
        },
      ],
    };
  }

• src/index.ts:269-292 (schema)

  The input schema defining parameters for the make_bot_speak tool, including required bot_id and text, and optional voice settings.

  inputSchema: {
    type: "object",
    properties: {
      bot_id: {
        type: "string",
        description: "ID of the bot that should speak",
      },
      text: {
        type: "string",
        description: "Text for the bot to speak",
      },
      voice_language_code: {
        type: "string",
        description: "Voice language code (optional, defaults to 'en-US')",
        default: "en-US",
      },
      voice_name: {
        type: "string",
        description: "Voice name (optional, defaults to 'en-US-Casual-K')",
        default: "en-US-Casual-K",
      },
    },
    required: ["bot_id", "text"],
  },

• src/index.ts:266-293 (registration)

  The tool registration entry in the ListTools response, specifying name, description, and input schema.

  {
    name: "make_bot_speak",
    description: "Make a bot speak text during a meeting using text-to-speech",
    inputSchema: {
      type: "object",
      properties: {
        bot_id: {
          type: "string",
          description: "ID of the bot that should speak",
        },
        text: {
          type: "string",
          description: "Text for the bot to speak",
        },
        voice_language_code: {
          type: "string",
          description: "Voice language code (optional, defaults to 'en-US')",
          default: "en-US",
        },
        voice_name: {
          type: "string",
          description: "Voice name (optional, defaults to 'en-US-Casual-K')",
          default: "en-US-Casual-K",
        },
      },
      required: ["bot_id", "text"],
    },
  },

• src/index.ts:410-411 (registration)

  The switch case in the CallToolRequest handler that dispatches calls to the makeBotSpeak handler.

  case "make_bot_speak":
    return await this.makeBotSpeak(args);