gemini

Control Gemini browser tabs and chats via MQTT WebSocket.

Updated Feb 28, 2026
One-click install
npx skills add https://github.com/heman1033p-create/heman1033p-create.github.io --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/heman1033p-create.github.io/tree/main/.gemini/skills/gemini
Command: npx skills add https://github.com/heman1033p-create/heman1033p-create.github.io --skill gemini-heman1033p-create

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Gemini users need direct, programmable control over the Gemini browser tab and related actions via MQTT WebSocket.

Core Features & Use Cases

  • Remote tab management: create_tab, list_tabs, focus_tab, inject_badge, and other commands to orchestrate Gemini tabs.
  • Chat and automation: send chat messages, transcribe YouTube videos, perform deep research, and switch models for different tasks.
  • Scripted workflows: includes ready-made scripts such as status, list-tabs, deep-research, full-smooth, and youtube-transcribe to automate common tasks.

Quick Start

Run the skill and issue commands such as '/gemini chat "Hello Gemini!"' to control the Gemini browser tab over MQTT WebSocket.

Frequently Asked Questions about gemini

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

FAQPage Schema
How do I automate Gemini chat messages using MQTT WebSocket?

You can automate Gemini chat messages by sending commands over an MQTT WebSocket connection to control the Gemini browser tab directly. This requires a Mosquitto broker with dual listeners on TCP 1883 and WebSocket 9001, and the Gemini Proxy Extension.

Can I transcribe YouTube videos automatically through Gemini?

Yes, you can transcribe YouTube videos by running the included youtube-transcribe script. This script automates the end-to-end flow of sending the video link to Gemini via MQTT WebSocket and retrieving the transcription.

What do I need to set up before controlling Gemini tabs via MQTT?

You need the Gemini Proxy Extension v2.8.8 or higher with its sidebar open, and a Mosquitto broker running dual listeners on TCP port 1883 for CLI scripts and WebSocket port 9001 for the browser connection.

How does remote tab management work for the Gemini browser?

Remote tab management works by issuing MQTT commands like create_tab, list_tabs, and focus_tab to orchestrate the browser. You can also inject UI badges and select different models for your automated tasks.

Are there ready-made scripts for automating Gemini research tasks?

Yes, the skill includes ready-made scripts such as deep-research, full-smooth, and youtube-transcribe to automate common workflows. These scripts handle sending prompts and managing tabs over the WebSocket.

Why is my MQTT WebSocket connection to the Gemini tab not working?

Your MQTT WebSocket connection may fail if the Mosquitto broker lacks the dual listeners on ports 1883 and 9001, or if the Gemini Proxy Extension sidebar is closed. Ensure the extension is v2.8.8 or higher.