Get the FREE Ultimate OpenClaw Setup Guide →

chatterboxio

A Model Context Protocol server implementation for ChatterBox, enabling AI agents to interact with online meetings and generate meeting summaries

Installation
Run this command in your terminal to add the MCP server to Claude Code.
Run in terminal:
Command
claude mcp add --transport stdio chatterboxio-chatterboxio-mcp-server npx -y @OverQuotaAI/chatterboxio-mcp-server \
  --env CHATTERBOX_API_KEY="<your_api_key_here>" \
  --env CHATTERBOX_API_ENDPOINT="https://api.chatter-box.io"

How to use

ChatterBox MCP Server provides AI agents with tools to join online meetings, capture transcripts and recordings, and generate concise meeting summaries. The server exposes commands that let agents join Zoom, Google Meet, or Microsoft Teams meetings, request meeting details, and produce summaries from transcripts. To use it, configure the required API endpoint and key for ChatterBox, then start the MCP server. Once running, you can invoke the available tools (joinMeeting, getMeetingInfo, summarizeMeeting) to enable meeting-related automation in your agent workflows.

How to install

Prerequisites:

  • Node.js 16+ (with npm or pnpm)
  • Access to the internet to install the MCP server package

Installation steps:

  1. Install dependencies from your project directory (choose npm or pnpm):
# Using npm
npm install

# Using pnpm
pnpm install
  1. Set up environment variables. Create a .env file in the project root with:
CHATTERBOX_API_ENDPOINT=https://api.chatter-box.io
CHATTERBOX_API_KEY=your_api_key_here
  1. Run the MCP server. You can start it via npm/pnpm scripts or directly with npx as shown below:
# Start using pnpm if a start script is defined
pnpm start

# Or run directly via npx (from anywhere with internet):
npx -y @OverQuotaAI/chatterboxio-mcp-server

Additional notes

Notes and tips:

  • Ensure the CHATTERBOX_API_ENDPOINT and CHATTERBOX_API_KEY are valid and active to access ChatterBox services.
  • The server supports joinMeeting (platform options: zoom, googlemeet, teams), getMeetingInfo (requires sessionId), and summarizeMeeting (requires a transcript).
  • If you run into rate limits, consider batching requests or implementing simple retry/backoff in your agent logic.
  • For debugging, you can use the MCP Inspector (npx @modelcontextprotocol/inspector) to inspect MCP protocol traffic and tool invocations.
  • If you deploy behind a firewall, ensure outbound access to the ChatterBox API endpoint is permitted.

Related MCP Servers

Sponsor this space

Reach thousands of developers