mcp -zoom-noauth

This MCP server provides Zoom capabilities without requiring the end user to perform a local authentication flow. It exposes a set of tools that interact with the Zoom API using OAuth credentials supplied at invocation time. The available tools are: zoom_refresh_token to refresh OAuth tokens, zoom_list_recordings to retrieve a user’s cloud recordings with pagination, zoom_get_recording_details to obtain detailed information about a specific recording, and zoom_get_meeting_transcript to fetch the transcript for a given meeting. To use these tools, run the server and invoke each tool by passing the required credentials and parameters as JSON payloads. Since this is a no-auth server, all credentials are supplied as tool arguments rather than stored on the server.

Prerequisites:

• Python 3.8+ and pip
• Git
• Optional: Docker if you want to run in containers

Install from source:

1. Clone the repository git clone https://github.com/peakmojo/mcp-server-zoom-noauth.git cd mcp-server-zoom-noauth

2. Install Python dependencies pip install -r requirements.txt

Run the server locally: python -m src.mcp_server_zoom_noauth.server

Optional: Run with Docker

1. Build the Docker image docker build -t mcp-server-zoom-noauth .

2. Run the Docker container (maps ports as needed) docker run -p 8000:8000 mcp-server-zoom-noauth

If you prefer Node (npm) setup, install the npm package and run the Node variant if provided by the repository package.json, otherwise use the Python route above.

Tips and caveats:

• This server is designed to operate without a local user authentication flow; all OAuth credentials are passed in at tool invocation time.
• Do not store access tokens or refresh tokens on the server; they are supplied with each tool call.
• Ensure the Zoom OAuth app has the required scopes for recordings access.
• When listing recordings, you can paginate using page_size and page_number.
• If you encounter API rate limits from Zoom, implement backoff in your client and retry logic in your tool usage.
• The server supports running in headless environments, which is useful for automation and CI.
• If you switch between Python and Node runtimes, ensure you use the correct invocation method for the chosen runtime.