gstack

Run persistent headless Chromium browser sessions for automated QA testing.

Updated Apr 13, 2026
One-click install
npx skills add https://github.com/legotec73/gstack --skill gstack-legotec73
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: gstack
Source: https://github.com/legotec73/gstack/tree/main
Command: npx skills add https://github.com/legotec73/gstack --skill gstack-legotec73

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

gstack enables AI agents to automate browser-based QA and site dogfooding with a persistent, headless Chromium daemon. It provides a reliable, ship-ready surface for Claude Code to drive real browser actions, capture evidence, and iterate on web flows without repeatedly spinning up new browser instances.

Core Features & Use Cases

  • Persistent browser sessions: cookies, tabs, and state survive across commands for end-to-end testing.
  • Rich UI interactions: navigate, click, fill forms, take screenshots, and run diffs to verify changes.
  • Evidence capture: annotated screenshots, diff outputs, and test artifacts to report bugs and verify deployments.
  • Quick-start testing workflows: validate user flows, test responsive layouts, and reproduce bugs with reproducible commands.

Quick Start

Start a gstack session and exercise a simple web flow with a single, explicit instruction.

Frequently Asked Questions about gstack

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

FAQPage Schema
How do I automate headless browser QA testing from Claude Code?

It provides a persistent headless Chromium daemon controlled from Claude Code to navigate, click, fill forms, take screenshots, and run diffs. This allows you to validate user flows and capture evidence without launching new browser instances.

Can I maintain cookies and browser state across multiple QA commands?

Yes, persistent browser sessions allow cookies, tabs, and state to survive across commands. This enables continuous end-to-end testing and site dogfooding workflows without losing authentication or session data between steps.

How do I capture evidence for bug reproduction during browser automation?

You capture evidence using annotated screenshots, diff outputs, and test artifacts generated by the CLI. Commands like snapshot, diff, and annotate visually document bugs and verify deployment changes.

Does this testing tool support responsive layout validation across environments?

Yes, the tool supports responsive testing across multiple pages and environments. You can validate user flows and test responsive layouts using reproducible commands within the persistent browser session.

What is needed to manage the headless Chromium daemon for testing?

You need a Bun-based CLI environment to start and manage the browser session. The standalone CLI loads a local server, manages sessions and cookies, and exposes commands to control the headless Chromium daemon.