google_meet

Join Google Meet calls, capture live captions, and optionally speak via OpenAI Realtime.

78|16|Updated Apr 23, 2026
One-click install
npx skills add https://github.com/sheawinkler/hermes-agent-ultra --skill google-meet-sheawinkler
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: google_meet
Source: https://github.com/sheawinkler/hermes-agent-ultra/tree/main/plugins/google_meet
Command: npx skills add https://github.com/sheawinkler/hermes-agent-ultra --skill google-meet-sheawinkler

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates joining Google Meet calls, captures live captions into a transcript, and optionally speaks in the meeting via OpenAI Realtime to assist with notes and follow-ups.

Core Features & Use Cases

  • Transcribes live captions from Google Meet and stores them in a transcript file for later review.
  • Real-time speaking: when enabled, the agent can speak in the meeting using a realtime OpenAI model and a virtual audio bridge.
  • Remote node support: run the bot on a separate machine (node host) and route commands from the gateway.
  • Safe, scoped operation: restricts to meet.google.com URLs, supports preflight installation, and provides status/follow-up capabilities.

Quick Start

Install prerequisites, authenticate, then join a Meet URL to start transcription; use meet join with mode=realtime if you want the agent to speak in the call.

Frequently Asked Questions about google_meet

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

FAQPage Schema
How do I automatically transcribe Google Meet calls into a transcript file?

To automatically transcribe Google Meet calls, you can use a Skill that joins the call via Playwright browser automation, captures live captions, and stores them in a local transcript file for later review.

Can an AI agent actively participate and speak in my Google Meet using OpenAI Realtime?

Yes, an AI agent can actively speak in Google Meet by utilizing a realtime OpenAI model and a virtual audio bridge to project speech into the ongoing call session.

Do I need Playwright to automate joining Google Meet for meeting notes?

Yes, Playwright is required for browser automation to join Google Meet calls, navigate the interface, and capture live captions for generating meeting notes and summaries.

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

Yes, you can run the Google Meet transcription bot on a separate remote machine by using node host support and routing commands from the gateway to keep execution isolated.

What is the best way to safely automate Google Meet without exposing other browser tabs?

The safest way to automate Google Meet is by using a Skill that restricts browser automation strictly to meet.google.com URLs, ensuring no other tabs or sessions are exposed.

How does real-time caption capture work for Google Meet summaries?

Real-time caption capture works by automating the Google Meet browser interface to extract live captions as they appear on screen, storing the text in a transcript file for summary generation.