ChatterBox MCP Server

The ChatterBox MCP Server bridges the gap between AI assistants and real‑time online meetings. By exposing a set of tools that can join Zoom, Google Meet, or Microsoft Teams sessions, capture transcripts and recordings, and generate concise summaries, it empowers developers to embed intelligent meeting assistants directly into their workflows. Instead of building custom integrations from scratch, teams can rely on this ready‑made MCP implementation to turn any meeting into a data source that AI agents can interrogate and act upon.

At its core, the server offers three primary capabilities. First, the joinMeeting tool connects an AI bot to a specified conference platform using credentials and meeting details supplied by the client. Once connected, it streams live audio and automatically transcribes the conversation, providing a rich textual record for downstream processing. Second, getMeetingInfo retrieves metadata and the accumulated transcript for a given session, allowing agents to query past meetings without manual intervention. Third, summarizeMeeting takes a raw transcript and produces a concise summary that highlights key decisions, action items, and discussion points—exactly the type of digest most participants need after a busy session.

Developers can integrate these tools into any MCP‑compatible client, such as Claude Desktop or other AI assistants. By invoking joinMeeting, an agent can attend a scheduled call, listen for specific keywords, or trigger follow‑up actions (e.g., posting a summary to Slack). After the meeting ends, summarizeMeeting can be called automatically to generate a report that is then emailed or stored in a knowledge base. This end‑to‑end flow removes manual note‑taking and ensures that critical information is captured reliably.

Real‑world scenarios include customer support teams that need instant post‑call summaries, product managers who want to track decisions across multiple sprint meetings, and compliance officers who require auditable transcripts. The server’s ability to handle multiple platforms out of the box and expose data through a consistent MCP interface makes it especially valuable in heterogeneous environments where meetings span Zoom, Google Meet, and Teams.

Unique advantages of this MCP server stem from its tight integration with the ChatterBox API, which handles authentication and meeting orchestration. The server also supports optional webhook callbacks for status updates, enabling reactive workflows that can trigger downstream processes as soon as a meeting starts or ends. Combined with the ease of installation via Smithery and the lightweight Node.js runtime, it offers a plug‑and‑play solution that accelerates AI‑powered meeting automation for developers and enterprises alike.