skills.rest
tropeks
tropeksCOMMUNITY
Explore all skills by this author →
GitHub

gstack

Drive a headless browser for QA testing and dogfooding tasks.

Updated Mar 26, 2026
Search Tags:#automation#qa#ai#claude#playwright#browser
One-click install
npx skills add https://github.com/tropeks/Vitali --skill gstack-tropeks
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: gstack
Source: https://github.com/tropeks/Vitali/tree/main/.claude/skills/gstack
Command: npx skills add https://github.com/tropeks/Vitali --skill gstack-tropeks

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires bun, playwright, claude, and includes scripts (resource) components.

What problem does it solve?

This skill enables Claude to drive a fast headless browser for QA testing and dogfooding tasks. It provides reliable tooling to navigate pages, interact with UI elements, verify state, and capture evidence during automated tests.

Core Features & Use Cases

  • Deterministic browser automation: Run deterministic commands to visit pages, click elements, fill forms, and capture results.
  • Evidence capture for QA: Diff pages before/after actions, take annotated screenshots, and generate proof of behavior.
  • Multi-scenario testing: Test layouts, forms, uploads, and dialogs across multiple pages or deployments with a persistent browser process.

Quick Start

Run a browse command to start testing a site.

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 with Claude?▼

Headless browser QA testing with Claude is enabled by driving deterministic commands to navigate pages, interact with UI elements, and capture evidence. The skill arbitrates commands with safety prompts and maintains a persistent local browser daemon.

Can I use Playwright to capture screenshots and diff UI evidence during automated testing?▼

Playwright can capture screenshots and diff UI evidence by running deterministic browser commands. The skill annotates screenshots and generates proof of behavior by comparing page states before and after interactions.

Does browser automation for QA require Bun to run locally?▼

Browser automation for QA requires Bun as a dependency alongside Playwright and Claude. Bun executes the scripts that start the lightweight, local browser daemon which persists state across multiple testing commands.

What is the best way to test responsive layouts across multiple deployments?▼

Testing responsive layouts across multiple deployments is best handled by running deterministic navigation and interaction commands against a persistent browser process. The skill supports multi-scenario testing for forms, uploads, and dialogs across various pages.

Why does command arbitration matter for AI-assisted browser automation?▼

Command arbitration in AI-assisted browser automation ensures safety by prompting before executing potentially destructive actions. This mechanism prevents unintended UI interactions while Claude drives the headless browser for QA and dogfooding tasks.

When should I not use a headless browser for QA testing?▼

A headless browser for QA testing should not be used when you need visual rendering validation or lack local execution environments for Bun and Playwright. It is designed for deterministic automated checks rather than manual visual review.

Related Skills

Explore More Software Engineering Skills→
imonmiimonmi

gstack

Automate browser QA with a persistent headless Chromium daemon.

Community
Advanced
MFS-codeMFS-code
8

browse

Run automated headless browser tests on user-provided URLs with screenshots and diffs.

Community
Advanced
drsapaevdrsapaev
2

webapp-testing

Test local web applications with Playwright Python scripts.

Community
Intermediate
NagyViktNagyVikt
2

agent-browser

Automate browser interactions and capture screenshots for UI verification.

Community
Advanced