How this skill is triggered — by the user, by Claude, or both
Slash command
/qa-auditor:qa-auditorThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
This skill provides the evaluation framework, data schema, and report generation tools for the `/qa-audit` command.
This skill provides the evaluation framework, data schema, and report generation tools for the /qa-audit command.
# Run a full audit with config
/qa-audit --config qa-audit.yaml
# Audit a single feature
/qa-audit --feature auth
# Auto-discovery mode (no config needed)
/qa-audit
File: references/evaluation-rubric.md
Defines the 6 quality dimensions scored 1-5:
Also defines issue severity levels: CRITICAL, HIGH, MEDIUM, LOW, INFO.
File: references/report-schema.md
JSON schema for audit-data.json — the structured data that feeds the HTML report. Sections: metadata, executive_summary, features, cross_cutting, issues, screenshots.
File: scripts/generate_report.py
Python script (stdlib only, no pip dependencies) that converts audit-data.json into a self-contained HTML report with embedded screenshots.
Usage:
python3 scripts/generate_report.py \
--input qa-audit-results/audit-data.json \
--output qa-audit-results/ \
--open
File: assets/report-template.html
The HTML/CSS/JS template used by the report generator. Features:
See examples/qa-audit.yaml for a complete reference. Minimum config:
project:
name: "My App"
environment:
base_url: "http://localhost:3000"
features:
- id: homepage
name: "Homepage"
priority: high
pages:
- url: "/"
name: "Home"
npx claudepluginhub andreherful/qa-auditor --plugin qa-auditorAutomates systematic web app QA testing with issue taxonomy, health scoring, and regression tracking. Supports browser automation via Playwright, Puppeteer, or Cypress.
QA tests web apps, native macOS/Electron apps, CLI tools, REPLs via browser/Playwright, computer-use screenshots/clicks, or tmux modes. Delivers health scores, bug fixes, ship-readiness reports.