skills.rest
heman1033p-create
heman1033p-createCOMMUNITY
Explore all skills by this author →
GitHub

gemini

Controls Gemini browser tabs via MQTT WebSocket commands for chat, research, and transcription.

Updated Feb 28, 2026
Search Tags:#browser automation#gemini#websocket#mqtt#deep research#chat automation#youtube transcription
One-click install
npx skills add https://github.com/heman1033p-create/cv-v2-kongphop --skill gemini-heman1033p-create
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: gemini
Source: https://github.com/heman1033p-create/cv-v2-kongphop/tree/main/.gemini/skills/gemini
Command: npx skills add https://github.com/heman1033p-create/cv-v2-kongphop --skill gemini-heman1033p-create

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires mqtt, and includes scripts (resource) components.

What problem does it solve? It enables programmatic control of Gemini browser tabs from the terminal, letting you send chat messages, run Deep Research, select models, and transcribe YouTube videos without manually interacting with the browser. ## Core Features & Use Cases - Tab Management: Create, list, focus, and target specific Gemini tabs by ID with badge injection for verification. - Chat & Research Automation: Send chat messages to specific tabs, select models (fast, thinking, pro), and trigger Deep Research workflows. - YouTube Transcription: Automate sending YouTube URLs to Gemini with structured transcription prompts. - Use Case: You want Gemini to transcribe a YouTube video hands-free. Run the youtube-transcribe script with the video URL, and it creates a new tab, injects a verification badge, and sends the transcription prompt automatically. ## Quick Start Ask the assistant to send a message to Gemini using the gemini skill, for example by saying "use gemini to chat Hello Gemini" after ensuring the Mosquitto broker and browser extension are running.

Frequently Asked Questions about gemini

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

FAQPage Schema
How do I send a chat message to Gemini from the command line?▼

Run the send-chat.ts script with your message as an argument using node --experimental-strip-types or Bun. It publishes a chat command to the claude/browser/command MQTT topic, and the browser extension types the message into the active Gemini tab.

How to automate Gemini Deep Research with MQTT?▼

Use the deep-research.ts script with your research topic as an argument. It creates a new Gemini tab, selects Deep Research mode, sends your prompt, and clicks the Start research button via MQTT commands.

What are the requirements for controlling Gemini via MQTT?▼

You need the Gemini Proxy browser extension v2.8.8 or later with its sidebar open, plus a Mosquitto broker listening on TCP port 1883 for CLI scripts and WebSocket port 9001 for the browser extension.

Why do MQTT commands to the Gemini extension time out?▼

Timeouts usually happen when using wrong topic names; the correct topics are claude/browser/command and claude/browser/response, not claude-browser-proxy/*. Also verify the extension sidebar is open and shows a green connected badge.

Can I target a specific Gemini tab when sending chat messages?▼

Yes, include the tabId parameter in the chat command payload. Use the list_tabs action or list-tabs.ts script to get available tab IDs, then pass the desired tabId to route the message to that exact tab.

Related Skills

Explore More Software Engineering Skills→
thiansitthiansit

gemini

Control Gemini browser tabs via MQTT WebSocket for messaging and automation.

Community
Intermediate
Soul-Brews-StudioSoul-Brews-Studio
116

gemini

Control Gemini browser tabs and chats via MQTT WebSocket commands.

Official
Advanced
alyamenfactoryalyamenfactory

ai-integrations-gemini

Access Gemini-compatible API endpoints via Replit AI Integrations proxy.

Community
Intermediate
hyperlane-xyzhyperlane-xyz
72

warp-fork

Fork a warp route from a local HTTP registry in the background.

Official
Advanced