google_meet

Join Google Meet URLs and capture live captions into a transcript.

1|Updated May 16, 2026
One-click install
npx skills add https://github.com/devMoez/titan --skill google-meet-devmoez
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: google_meet
Source: https://github.com/devMoez/titan/tree/main/plugins/google_meet
Command: npx skills add https://github.com/devMoez/titan --skill google-meet-devmoez

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires playwright, websockets, and includes assets (resource) components.

What problem does it solve?

It eliminates the manual effort of joining Google Meet, listening for key discussion points, and turning live captions into a usable transcript and follow-up work.

Core Features & Use Cases

  • Live caption transcription: Joins a specified Google Meet URL and scrapes live captions into a transcript file you can read and summarize.
  • Optional in-call speaking (realtime): In realtime mode, the agent can speak into the meeting using OpenAI Realtime, coordinated via a virtual audio bridge.
  • Local or remote node execution: Run the bot on the gateway machine (default) or offload to a remote node host (e.g., a user’s Mac with a signed-in Chrome profile) for practical audio/browser setup.
  • Use cases: Standup note-taking, meeting summaries with action items, responding in-call when requested, and generating recap messages after the meeting ends.

Quick Start

Use this skill when the user says: "Join my Meet at https://meet.google.com/<code> and take notes."

Frequently Asked Questions about google_meet

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate Google Meet transcription and meeting notes?

To automate Google Meet transcription, a headless Playwright session joins the specified meeting URL and captures live captions into a persistent transcript file for note-taking and action-item generation.

Can a bot speak in a Google Meet call in realtime?

An agent can speak into a Google Meet call in realtime mode by using OpenAI Realtime, coordinating speech output through a virtual audio bridge while still tracking live caption transcripts.

Do I need Playwright and websockets to run a Google Meet note-taking bot?

Yes, Playwright is required to manage the headless browser session for joining meetings and scraping captions, while websockets enable optional remote node execution via JSON-over-WebSocket RPC with token-gated requests.

How do I capture live captions from a Google Meet URL?

Capturing live captions requires validating the meet.google.com URL pattern, launching a headless Playwright browser to join the call, and tracking meeting state while persisting caption text into a transcript file.

Can I run a Google Meet transcription bot on a remote node?

You can offload the Google Meet transcription bot to a remote node host, such as a Mac with a signed-in Chrome profile, using JSON-over-WebSocket RPC with token-gated requests to manage browser and audio setup.

What is the best way to generate action items from meeting notes?

Generating action items from meeting notes is best handled by capturing live voice captions into a persistent transcript during the call, then processing that transcript text to extract tasks and summarize key discussion points.