gstack

Automates browser-based QA tasks with a headless Chromium daemon.

1|Updated Mar 28, 2026
One-click install
npx skills add https://github.com/shhubbh/flowstate --skill gstack-shhubbh
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: gstack
Source: https://github.com/shhubbh/flowstate/tree/main/.claude/skills/gstack
Command: npx skills add https://github.com/shhubbh/flowstate --skill gstack-shhubbh

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This tool accelerates QA workflows by providing a persistent, headless browser controlled by Claude Code, enabling deterministic testing and automated bug reproduction without manual setup.

Core Features & Use Cases

  • Headless browser automation: Navigate, click, fill, read content, and capture screenshots for test evidence.
  • Diffs and annotated reports: Produce before/after diffs and annotated visuals to validate changes in PRs.
  • Dogfooding and verification: Validate deployments in staging, reproduce user flows, and document issues with reproducible steps.

Quick Start

Start by launching the browse daemon and running a simple QA workflow to verify a page.

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 tasks and capture screenshots for PR evidence?

Automating headless browser QA tasks is done by controlling a persistent Chromium daemon to navigate, click, fill forms, and capture screenshots. This approach provides deterministic test execution and generates visual evidence directly usable in pull requests.

Can I reproduce bugs automatically using a headless Chromium workflow?

Reproducing bugs automatically with a headless Chromium workflow allows you to script user flows across multiple pages. By navigating and reading state, you can replicate issue conditions deterministically without manual browser setup.

What is the best way to generate visual diffs for staging deployments?

Generating visual diffs for staging deployments involves capturing before and after screenshots during automated browser workflows. The tool produces annotated visuals to validate changes and document issues with reproducible steps.

Do I need a local browse server to run end-to-end browser automation?

A local browse server is required to run end-to-end browser automation with this tool. The server hosts the persistent headless Chromium daemon that executes commands like goto, snapshot, and diff for your workflows.

How does dogfooding work with automated browser navigation and state diffing?

Dogfooding with automated browser navigation uses a headless daemon to validate deployments and reproduce user flows. By executing commands to fill and snapshot state, you generate reproducible steps to document issues found during internal testing.