tab-agent

Control a live Chrome session from the CLI to automate browsing tasks.

8|1|Updated Jan 30, 2026
One-click install
npx skills add https://github.com/DrHB/tab-agent --skill tab-agent
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: tab-agent
Source: https://github.com/DrHB/tab-agent/tree/main/skills/claude-code/tab-agent
Command: npx skills add https://github.com/DrHB/tab-agent --skill tab-agent

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill enables you to control a live Chrome browser session from the command line, letting you automate navigation, input, and interactions without launching headless instances or re-authenticating.

Core Features & Use Cases

  • CLI-based browser automation: navigate, click, type, scroll, hover, and extract data from an active tab.
  • Attach to your real session via the extension icon (green = active) to work with your actual logins and cookies.
  • Snapshot-driven element refs for reliable interaction and repeatable scripts.
  • Use cases include rapid form filling, automated testing-like flows, data gathering from sites where you are already logged in, and personal automation tasks.

Quick Start

Start the Tab Agent service and begin issuing commands like navigate and snapshot to control your active browser tab.

Frequently Asked Questions about tab-agent

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

FAQPage Schema
How do I automate browser tasks from the CLI using my existing Chrome session?

You can automate browser tasks from the CLI by attaching to an active Chrome session via the extension. This allows you to execute commands for navigation, clicking, typing, and data extraction using your actual logins and cookies without re-authenticating.

What is CLI-based browser automation for a live Chrome session?

CLI-based browser automation controls a real, active Chrome tab directly from the command line. It enables deterministic actions like navigate, click, type, and snapshot to perform repetitive interactions and web navigation without launching headless browser instances.

Can I extract data from websites where I am already logged in via command line?

Yes, you can extract data from logged-in sites. By attaching the extension to your active Chrome session, the CLI commands operate with your existing cookies and authentication state, allowing seamless data gathering and form filling.

How do I ensure reliable element interaction during web automation scripting?

You ensure reliable interaction by using snapshot-driven element references. Taking a snapshot provides deterministic element refs for clicking and typing, which creates repeatable scripts and stable automation flows across various websites.

Does this browser automation approach require a headless browser setup?

No, it does not require a headless browser setup. The tool attaches directly to your live, active Chrome session via an extension, bypassing the need to launch headless instances or manage separate authentication profiles for your automation workflows.

What are the limitations of controlling Chrome from the command line?

The primary limitation is the dependency on an active, live Chrome session and its connected extension. It is designed for deterministic actions on a single active tab rather than massive parallel scraping, requiring the browser session to remain open during task execution.