gstack

Automate browser navigation, interaction, and snapshot validation via a Bun/Playwright daemon.

Updated Nov 11, 2025

Search Tags:#automation #qa #cli #diff #playwright #browser #screenshot

Ready to run

npx skills add https://github.com/rshelekhov/dotfiles --skill gstack-rshelekhov

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: gstack
Source: https://github.com/rshelekhov/dotfiles/tree/main/claude/skills/gstack
Command: npx skills add https://github.com/rshelekhov/dotfiles --skill gstack-rshelekhov

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires playwright, diff.

What problem does it solve?

gstack provides a fast, lightweight, headless browser automation layer for Claude Code, enabling teams to test, QA, and dogfood apps with reliable state, fast feedback, and actionable evidence.

Core Features & Use Cases

Navigate pages, interact with elements, and generate diffs plus annotated screenshots to verify changes and communicate findings.
Maintain per-workspace browser state with persistent cookies and tabs for repeatable tests across sessions.
Use in CLI-driven workflows for QA, design reviews, and automated testing pipelines.

Quick Start

Install dependencies and run the CLI to start a headless browser session, then issue commands to navigate, interact, and capture evidence.

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 testing with Claude Code?▼

Headless browser automation for Claude Code workflows is handled by running a persistent Bun/Playwright daemon that executes bash-based commands to navigate pages, interact with elements, and capture screenshots for automated testing.

Can I generate visual diffs and annotated screenshots during browser automation?▼

Yes, visual diff-driven validation generates annotated screenshots during browser automation, enabling you to verify changes and communicate findings with actionable evidence for design reviews and QA.

Does Playwright support persistent browser state for repeatable QA sessions?▼

Playwright supports persistent browser state through per-workspace isolation, maintaining cookies and tabs across sessions to ensure repeatable tests and reliable state for automated testing pipelines.

What's the best way to run end-to-end browser automation in a CLI workflow?▼

End-to-end browser automation in CLI-driven workflows is best handled by a persistent daemon that applies bash-based commands to live sites, providing fast feedback and reliable state for QA and automated testing.

How do I interact with browser elements and capture snapshots using a CLI?▼

You interact with browser elements and capture snapshots using bash-based commands issued to a headless browser session, enabling fast actions like navigate, interact, and snapshot across live sites.

Do I need Bun to run headless browser automation for QA?▼

Bun is used to maintain the persistent Playwright daemon for headless browser automation, providing fast feedback and reliable state for QA, design reviews, and automated testing across live sites.