Google Meet & Calendar MCP Server

A production-ready Model Context Protocol (MCP) server for Google Meet and Google Calendar. Connect any AI agent to Google Calendar & Meet — create meetings with Meet links, manage events, browse calendars, access recordings, transcripts, and participant data.

This is an independent, generalized MCP server. It is not tied to any specific project and can be connected to any AI agent that supports MCP (Claude Desktop, Cursor, WorkOS, or custom agents).

Features

• 📅 Calendar Events: Create, read, update, delete, and list events
• 🎥 Meetings: Create Google Meet meetings with auto-generated links
• 👥 Attendees: Add attendees to events
• 📋 Calendars: List all accessible calendars
• 🎬 Recordings: List and download meeting recordings (Meet REST API v2)
• 📝 Transcripts: Access meeting transcripts (Meet REST API v2)
• 👤 Participants: View meeting participant history (Meet REST API v2)
• 📊 Conference Records: Browse past meeting records

Installation

Recommended — no pre-install required (uses uv):

# uv manages a temporary isolated environment automatically
uvx --from workos-gmeet-mcp-server gmeet-mcp-server

Or install permanently with pip:

pip install workos-gmeet-mcp-server

Or install from source:

git clone https://github.com/workos/workos-gmeet-mcp-server
cd workos-gmeet-mcp-server
pip install -e .

Quick Start

export GOOGLE_CLIENT_ID="your-client-id.apps.googleusercontent.com"
export GOOGLE_CLIENT_SECRET="GOCSPX-your-secret"
export GOOGLE_REFRESH_TOKEN="1//your-refresh-token"

# Recommended: run via uvx (no install needed)
uvx --from workos-gmeet-mcp-server gmeet-mcp-server

# Or if installed via pip
gmeet-mcp-server

# Or run as a Python module
python -m gmeet_mcp_server

Configuration

Environment Variables

Variable	Required	Description
GOOGLE_CLIENT_ID	✅ Yes	OAuth2 Client ID from Google Cloud Console
GOOGLE_CLIENT_SECRET	✅ Yes	OAuth2 Client Secret
GOOGLE_REFRESH_TOKEN	✅ Yes	OAuth2 Refresh Token (generated once, used to auto-refresh access tokens)

Getting Google OAuth2 Credentials

1. Create a Google Cloud Project

1. Go to Google Cloud Console
2. Create a new project (or select existing)
3. Enable these APIs under APIs & Services → Library:
   • Google Calendar API
   • Google Meet REST API (for recordings/transcripts)
   • Google Drive API (for downloading recordings)

2. Configure OAuth Consent Screen

1. Go to APIs & Services → OAuth consent screen
2. Choose External user type
3. Fill in app name and email
4. Add scopes:
   • https://www.googleapis.com/auth/calendar
   • https://www.googleapis.com/auth/meetings.space.readonly
   • https://www.googleapis.com/auth/drive.readonly

3. Create OAuth2 Credentials

1. Go to APIs & Services → Credentials
2. Click Create Credentials → OAuth client ID
3. Application type: Desktop app
4. Note down the Client ID and Client Secret

4. Generate a Refresh Token

Option A — Using Google OAuth Playground:

1. Go to OAuth 2.0 Playground
2. Click ⚙️ Settings → check Use your own OAuth credentials → enter your Client ID & Secret
3. In Step 1, add scopes:
   • https://www.googleapis.com/auth/calendar
   • https://www.googleapis.com/auth/meetings.space.readonly
   • https://www.googleapis.com/auth/drive.readonly
4. Click Authorize APIs → sign in → grant access
5. In Step 2, click Exchange authorization code for tokens
6. Copy the Refresh Token

Option B — Using the included helper script:

python MCP_servers/get_google_token.py \
  --client-id "YOUR_CLIENT_ID" \
  --client-secret "YOUR_CLIENT_SECRET"

Connecting to AI Agents

Note: Use uvx --from workos-gmeet-mcp-server gmeet-mcp-server in all configs below. The --from flag is needed because the PyPI package name (workos-gmeet-mcp-server) differs from the CLI entry point (gmeet-mcp-server). You must have uv installed (brew install uv on macOS).

Claude Desktop

Edit ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "gmeet": {
      "command": "uvx",
      "args": ["--from", "workos-gmeet-mcp-server", "gmeet-mcp-server"],
      "env": {
        "GOOGLE_CLIENT_ID": "your-client-id.apps.googleusercontent.com",
        "GOOGLE_CLIENT_SECRET": "GOCSPX-your-secret",
        "GOOGLE_REFRESH_TOKEN": "1//your-refresh-token"
      }
    }
  }
}

Alternative — if you have installed via pip install workos-gmeet-mcp-server and prefer not to use uvx, use the full absolute path to the binary instead:

{ "command": "/path/to/your/bin/gmeet-mcp-server" }

Find the path with: which gmeet-mcp-server

Cursor

Edit ~/.cursor/mcp.json (global) or .cursor/mcp.json (project):

{
  "mcpServers": {
    "gmeet": {
      "command": "uvx",
      "args": ["--from", "workos-gmeet-mcp-server", "gmeet-mcp-server"],
      "env": {
        "GOOGLE_CLIENT_ID": "your-client-id.apps.googleusercontent.com",
        "GOOGLE_CLIENT_SECRET": "GOCSPX-your-secret",
        "GOOGLE_REFRESH_TOKEN": "1//your-refresh-token"
      }
    }
  }
}

VS Code (GitHub Copilot / MCP extension)

Edit .vscode/mcp.json in your project:

{
  "servers": {
    "gmeet": {
      "type": "stdio",
      "command": "uvx",
      "args": ["--from", "workos-gmeet-mcp-server", "gmeet-mcp-server"],
      "env": {
        "GOOGLE_CLIENT_ID": "${input:googleClientId}",
        "GOOGLE_CLIENT_SECRET": "${input:googleClientSecret}",
        "GOOGLE_REFRESH_TOKEN": "${input:googleRefreshToken}"
      }
    }
  },
  "inputs": [
    {
      "id": "googleClientId",
      "type": "promptString",
      "description": "Google OAuth2 Client ID"
    },
    {
      "id": "googleClientSecret",
      "type": "promptString",
      "description": "Google OAuth2 Client Secret",
      "password": true
    },
    {
      "id": "googleRefreshToken",
      "type": "promptString",
      "description": "Google OAuth2 Refresh Token",
      "password": true
    }
  ]
}

WorkOS / Custom Agents

Add to .mcp.json in your project root:

{
  "mcpServers": {
    "gmeet": {
      "transport": "stdio",
      "command": "uvx",
      "args": ["--from", "workos-gmeet-mcp-server", "gmeet-mcp-server"],
      "env": {
        "GOOGLE_CLIENT_ID": "${GOOGLE_CLIENT_ID}",
        "GOOGLE_CLIENT_SECRET": "${GOOGLE_CLIENT_SECRET}",
        "GOOGLE_REFRESH_TOKEN": "${GOOGLE_REFRESH_TOKEN}"
      }
    }
  }
}

Available Tools (14)

Calendar Events

Tool	Description
gmeet_list_events	List upcoming calendar events
gmeet_create_event	Create a calendar event (no Meet link)
gmeet_get_event	Get details of a specific event
gmeet_update_event	Update an existing event
gmeet_delete_event	Delete a calendar event

Meetings

Tool	Description
gmeet_create_meeting	Create a calendar event WITH a Google Meet link
gmeet_add_attendee	Add an attendee to an existing event

Calendars

Tool	Description
gmeet_list_calendars	List all accessible calendars

Recordings & Transcripts (Meet REST API v2)

Tool	Description
gmeet_list_conference_records	Browse past meeting records
gmeet_get_meeting_participants	View participant history for a meeting
gmeet_list_recordings	List recordings for a meeting
gmeet_list_transcripts	List transcripts for a meeting
gmeet_download_recording	Download a meeting recording via Google Drive
gmeet_get_transcript_content	Get transcript text content via Google Docs

Development

git clone https://github.com/workos/workos-gmeet-mcp-server
cd workos-gmeet-mcp-server
pip install -e .

# Run tests
pytest

# Run the server locally
GOOGLE_CLIENT_ID=... GOOGLE_CLIENT_SECRET=... GOOGLE_REFRESH_TOKEN=... python -m gmeet_mcp_server

Publishing

To PyPI

pip install build twine
rm -rf dist/ build/ src/*.egg-info
python -m build
pip install "packaging>=24.2"   # required for Metadata 2.4 support in twine
twine check dist/*              # validate before uploading
twine upload dist/*

To MCP Registry

mcp-publisher login github
mcp-publisher publish

License

MIT — see LICENSE for details.