One-click install
npx skills add https://github.com/MFS-code/gstack --skill qa-mfs-code
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: qa
Source: https://github.com/MFS-code/gstack/tree/main/qa
Command: npx skills add https://github.com/MFS-code/gstack --skill qa-mfs-code

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

It helps you systematically QA a web application by finding real bugs in the UI and flows, fixing them in source code with atomic commits, and re-verifying that the fixes didn’t break anything else.

Core Features & Use Cases

  • Manual-trigger QA with tiered thoroughness: run Quick for only critical/high fixes, Standard for typical severity coverage, or Exhaustive for deeper issues including low/cosmetic problems.
  • Evidence-rich reporting: produces before/after health scores plus screenshots and fix evidence, then summarizes ship-readiness.
  • Diff-aware verification: when no URL is provided on a feature branch, it scopes testing based on what changed and exercises the most likely affected routes/pages.
  • Optional report-only mode: supports a mode to generate findings without making code fixes.

Quick Start

Run qa against your app by invoking /qa and tell it the target URL you want tested.

Frequently Asked Questions about qa

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

FAQPage Schema
How do I run end-to-end QA testing on a local web application and fix discovered bugs?

Run end-to-end QA testing by triggering the skill with your target URL, and it will discover UI bugs, fix the source code with atomic commits, and re-verify the changes. This ensures your web application's interactive flows work without introducing regressions.

What is diff-scoped page testing and when do I need it for regression checks?

Diff-scoped page testing evaluates only the routes and pages affected by recent code changes on a feature branch. You need it for regression-style checks when no specific URL is provided, ensuring targeted verification of interactive UI flows without testing the entire application.

Can I generate a QA bug report with screenshot evidence without modifying my source code?

Yes, you can use the optional report-only mode to generate findings with screenshot evidence and health scores without applying code fixes. This allows you to review identified UI bugs and severity tiers before committing any changes.

Does this QA process support different levels of bug fixing thoroughness?

Yes, the QA process supports tiered thoroughness levels: Quick for critical fixes, Standard for typical severity coverage, and Exhaustive for identifying deeper issues including low-severity cosmetic problems. You manually trigger the desired level for your web application testing.

How does health scoring verify that UI fixes do not break existing application functionality?

Health scoring provides before and after metrics to verify that UI fixes do not break existing functionality. The skill performs safe atomic commits and re-checks interactive flows, producing structured reports with screenshot evidence to summarize ship-readiness.