ChatterBox MCP Server is a Model Context Protocol (MCP) server implementation for ChatterBox, designed to enable AI agents to interact with online meetings and generate meeting summaries.
Purpose: This server provides tools for AI agents to enable them to join online meetings, capture transcripts and recordings, and generate meeting summaries.
Main capabilities:
- Join online meetings: Supports Zoom, Google Meet, or Microsoft Teams.
- Capture meeting transcripts and audio recordings.
- Generate meeting summaries: Generate concise summaries based on meeting transcript content.
Usage:
- Installation: chatterboxio-mcp-server can be automatically installed for Claude Desktop via Smithery, or dependencies can be installed manually using npm or pnpm.
- Obtain an API key: Register for free on the ChatterBox official website; after registration, you will receive an API endpoint and key.
- Environment configuration: Create a .env file in the root directory and set the corresponding variables.
- After starting the server, the following tools are available:
- Join a meeting and capture transcripts and audio recordings: Parameters include platform (string, optional, "zoom", "googlemeet", or "teams"), meetingId (string, meeting ID), meetingPassword (string, optional, meeting password or passcode), botName (string, bot name), webhookUrl (string, optional, URL to receive meeting status webhook events).
- Get meeting information: The parameter is sessionId (string, session ID), returning meeting information including transcripts and recordings.
- Generate a meeting summary: The parameter is transcript (string, meeting transcript content), generating a concise summary.
- Debugging: MCP Inspector can be used to debug the MCP server.
Use cases: Scenarios that require AI agents to automatically participate in online meetings, record meeting content, and generate summaries.
Limitations: Specific environment variable names, startup commands, etc. are not provided in the materials; pricing and billing models are not yet provided.
