gstack

Orchestrate Markdown-based skills for browser automation and QA workflows.

Updated Mar 31, 2026
One-click install
npx skills add https://github.com/huichen/gstack --skill gstack-huichen
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: gstack
Source: https://github.com/huichen/gstack/tree/main
Command: npx skills add https://github.com/huichen/gstack --skill gstack-huichen

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

gstack provides a unified framework that turns Claude Code prompts into an engineering workflow by defining a library of Markdown-based skills and a browser automation layer, enabling teams to ship faster with AI agents. It centralizes QA, code reviews, design reviews, and design-to-code pipelines into repeatable, loadable units.

Core Features & Use Cases

  • A library of self-contained skills loaded from SKILL.md frontmatter, enabling structured agent workflows.
  • Built-in browser automation with a persistent headless daemon for rapid end-to-end testing and debugging.
  • Workflow pipelines that chain CEO reviews, architecture planning, design reviews, QA, ship, and doc updates to reduce context switching.
  • Use cases include automating PR reviews, QA testing, design exploration, and auto-generating design docs.

Quick Start

Install gstack, then run /office-hours to start diagnosing your product idea and load the skill suite.

Frequently Asked Questions about gstack

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

FAQPage Schema
How do I automate QA testing and PR reviews with an AI workflow?

You can automate QA testing and PR reviews by orchestrating a library of Markdown-based skills loaded from frontmatter, enabling an AI agent to execute structured end-to-end engineering workflows.

What is a Markdown-based skill library for AI engineering workflows?

A Markdown-based skill library is a collection of self-contained units loaded from SKILL.md frontmatter that define repeatable agent workflows for tasks like code reviews, design exploration, and documentation updates.

Does AI browser automation require a specific runtime environment?

AI browser automation requires a local Bun runtime and a Playwright-backed browser daemon to enable persistent, scriptable headless browsing for rapid end-to-end testing and debugging.

Can I chain multiple engineering tasks like architecture planning and ship flows?

Yes, you can chain multiple engineering tasks by defining workflow pipelines that sequence architecture planning, design reviews, QA, ship flows, and doc updates to reduce context switching.

What is the best way to start diagnosing a product idea using an AI engineering agent?

The best way to start diagnosing a product idea is to run an office-hours command that loads the skill suite and initializes the AI-driven engineering workflow framework.

Why use frontmatter-driven SKILL.md files for AI workflow automation?

Frontmatter-driven SKILL.md files are used to load self-contained skills into the AI agent, providing structured metadata that enables repeatable, scriptable workflows across browser automation and QA pipelines.