Zoom Transcript MCP Server

What is Zoom Transcript MCP Server?

An MCP (Model Context Protocol) server for interacting with Zoom Cloud Recording transcripts. It allows you to list, download, search, and manage your Zoom meeting transcripts through a structured interface, intended for developers using MCP-compatible assistants.

How to use Zoom Transcript MCP Server?

Install dependencies with npm install, build with npm run build, then configure via a .env file or MCP settings with Zoom OAuth credentials (Account ID, Client ID, Client Secret) and an optional transcripts directory. Invoke tools like list_meetings, download_transcript, get_recent_transcripts, and search_transcripts through an MCP client.

Key features of Zoom Transcript MCP Server

• List all available Zoom meetings with recordings
• Download transcripts by meeting ID or UUID
• Automatically download recent meeting transcripts
• Search across all downloaded transcripts
• Transcripts organized in a monthly folder structure

Use cases of Zoom Transcript MCP Server

• Retrieve and review transcripts from past Zoom meetings
• Search for specific topics or keywords across multiple transcripts
• Automatically archive recent meeting transcripts locally
• Integrate transcript access into AI assistants for summarization or analysis

FAQ from Zoom Transcript MCP Server

What prerequisites are needed to run the server?

Node.js v16 or higher, a Zoom account with Cloud Recording enabled, and a Zoom Server-to-Server OAuth app with cloud_recording:read:list_account_recordings:admin, cloud_recording:read:recording:admin, and cloud_recording:read:list_user_recordings:admin scopes.

How do I obtain OAuth credentials from Zoom?

Go to the Zoom App Marketplace, click "Develop" > "Build App", choose "Server-to-Server OAuth", fill in required info, add the required scopes, save and activate, then note your Account ID, Client ID, and Client Secret.

Where are downloaded transcripts stored?

Transcripts are saved locally in a structured directory (./transcripts by default, configurable via TRANSCRIPTS_DIR). Each month has a folder (e.g., YYYY-MM) containing .vtt transcript files and a metadata/ subfolder with corresponding JSON metadata.

What tools does the server provide?

Four tools: list_meetings (with optional date range and participant filter), download_transcript (by meeting ID/UUID), get_recent_transcripts (with optional count), and search_transcripts (with query and optional date range).

How do I configure the server through my MCP settings?

In your MCP settings file, add an entry under mcpServers with command "node", args pointing to the built index.js, and environment variables for ZOOM_ACCOUNT_ID, ZOOM_CLIENT_ID, ZOOM_CLIENT_SECRET, and optionally TRANSCRIPTS_DIR.