google_meet

Joins Google Meet calls with a Playwright bot to transcribe captions and speak via OpenAI Realtime.

Updated May 3, 2026
One-click install
npx skills add https://github.com/80portisfound/vibe-learning --skill google-meet-80portisfound
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: google_meet
Source: https://github.com/80portisfound/vibe-learning/tree/main/packages/hermes/plugins/google_meet
Command: npx skills add https://github.com/80portisfound/vibe-learning --skill google-meet-80portisfound

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires playwright, websockets.

What problem does it solve? Attending and documenting Google Meet calls manually is time-consuming, and automating a bot that can both listen and speak in a call requires solving browser automation, caption scraping, and virtual audio routing. This Skill lets an agent join a Meet URL, scrape live captions into a transcript, optionally speak in the call, and handle follow-up work afterwards. ## Core Features & Use Cases - Transcribe mode: A headless Chromium bot joins the meeting, enables live captions, and scrapes them into a timestamped transcript file via a DOM MutationObserver. - Realtime voice mode: The bot speaks queued text into the call through OpenAI Realtime, routed into Chrome's fake microphone via a PulseAudio null-sink (Linux) or BlackHole (macOS), with best-effort barge-in when humans talk. - Remote node hosts: Run the bot on a different machine than the gateway (e.g., a Mac with a signed-in Chrome) over a token-authenticated WebSocket connection. - Use Case: Ask the agent to sit in on your standup, take notes, then summarize the transcript and send follow-ups to the team after the call ends. ## Quick Start Ask the agent to join your Google Meet URL and take notes on the meeting, then summarize the transcript when the call ends.

Frequently Asked Questions about google_meet

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

FAQPage Schema
How do I make a bot join a Google Meet call and take notes?▼

Call meet_join with the meet.google.com URL and the bot joins via headless Chromium, enables live captions, and scrapes them into a transcript file. Poll with meet_status and read captions with meet_transcript, then call meet_leave when done.

How can an AI agent speak in a Google Meet call?▼

Join with mode='realtime' and call meet_say with the text to speak. The bot sends the text to OpenAI Realtime and streams the generated PCM audio into Chrome's fake microphone through a PulseAudio null-sink on Linux or BlackHole on macOS.

Does the Google Meet bot work on Windows?▼

No, the plugin only supports Linux and macOS. The register function refuses to load on Windows because the realtime audio routing path is untested there and guest-join Chromium behavior is flaky.

Can the Meet bot run on a different machine than the agent?▼

Yes, run 'hermes meet node run' on the machine with a signed-in Chrome, then approve it on the gateway with 'hermes meet node approve'. Pass node='<name>' to any meet tool to operate the remote bot over a token-authenticated WebSocket.

Why is the Meet bot stuck in the lobby?▼

Guest-mode joins wait in the lobby until a host admits them, and the bot leaves with leaveReason 'lobby_timeout' after 5 minutes by default. Run 'hermes meet auth' to save a signed-in Google session and skip the guest lobby.

What are the limitations of Google Meet caption scraping?▼

Transcripts are only as good as Google Meet's built-in live captions, which are English-biased and lossy on overlapping speakers. The approach avoids STT API costs but depends on Meet's DOM structure and caption availability.