gstack

Provision a persistent headless Chromium daemon for browser-based QA testing.

Updated Apr 7, 2026
One-click install
npx skills add https://github.com/zz8011/harnesstest --skill gstack-zz8011
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: gstack
Source: https://github.com/zz8011/harnesstest/tree/main/.claude/skills/gstack
Command: npx skills add https://github.com/zz8011/harnesstest --skill gstack-zz8011

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires playwright, @anthropic-ai/sdk, diff, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

gstack provides a fast, persistent headless browser daemon to empower QA testing, bug reproduction, and in-conversation web dogfooding inside Claude Code workflows.

Core Features & Use Cases

  • Persistent Chromium daemon with sub-second command latency for rapid QA iterations.
  • Navigate, interact, extract content, take annotated screenshots, and diff changes to verify behavior.
  • Dogfood user flows and validate deployments by running real interactions in a real browser.

Quick Start

Start the browse daemon and begin testing a site with Claude Code using /browse.

Frequently Asked Questions about gstack

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

FAQPage Schema
How do I automate browser-based QA testing inside Claude Code?

Automate browser-based QA testing using a persistent headless Chromium daemon that handles navigation, interaction, and content extraction. It coordinates these actions within Claude Code workflows to execute rapid iterations and verify web application behavior.

Can I run Playwright tests directly within my Claude Code workflow?

Yes, Playwright tests can run directly within Claude Code workflows. The system provisions a persistent headless Chromium daemon using Playwright to perform navigation, take annotated screenshots, and collect evidence with sub-second command latency.

What is the best way to dogfood user flows and validate web deployments?

Dogfooding user flows is best achieved by running real interactions in a persistent headless browser. This approach validates deployments by executing navigation and interaction sequences, then capturing visual diffs and screenshots as evidence of behavior.

How do I capture screenshots and diff visual changes during headless browser testing?

Capture screenshots and diff visual changes during testing by utilizing the daemon's built-in content extraction and visual diff capabilities. It automatically records annotated screenshots and compares changes to verify application behavior within your workflow.

Does headless browser testing require a persistent daemon for rapid QA iterations?

Yes, a persistent daemon is required for rapid QA iterations during headless browser testing. It provides sub-second command latency, allowing you to quickly navigate, interact, and extract content without the overhead of restarting the browser.

Why does my headless browser testing slow down during rapid QA iterations?

Headless browser testing slows down during rapid QA iterations when a persistent daemon is not used. Provisioning a persistent Chromium daemon eliminates browser restart overhead, maintaining sub-second command latency for continuous navigation and interaction.