npx claudepluginhub neonwatty/qa-skillsQA testing pipeline with 6 personas (smoke, UX, adversarial, security, performance, mobile) — generate workflow docs, convert to Playwright E2E tests, run interactively or in CI. Supports quantified UX scoring with before/after binary scorecards, Next.js performance profiling, and mobile UX auditing against iOS HIG and Material Design 3 standards.
Production-ready workflow orchestration with 79 focused plugins, 184 specialized agents, and 150 skills - optimized for granular installation and minimal token usage
Curated collection of 141 specialized Claude Code subagents organized into 10 focused categories
Directory of popular Claude Code extensions including development tools, productivity plugins, and MCP integrations
QA testing pipeline for Claude Code — generate workflow docs, convert to Playwright E2E tests, run interactively or in CI, and audit apps with six specialized QA agents. Supports desktop, mobile, and multi-user flows with profile-based authentication.
Full walkthrough: Claude Code Browser Testing and iOS Automation with MCP Workflows
claude plugin marketplace add neonwatty/qa-skills
claude plugin install qa-skills@neonwatty-qa
# 1. Install Playwright CLI (one-time, global)
npm install -g @playwright/cli@latest
playwright-cli install
# 2. Create auth profiles for your app (one-time per project)
/setup-profiles
/setup-profiles opens a headed browser for each user role. You log in manually (handles OAuth, 2FA, etc.) and the session state is saved to .playwright/profiles/. All generators, agents, and the runner load these profiles automatically.
┌→ Converters → .spec.ts → CI (GitHub Actions)
/setup-profiles → Generators → workflow ─┤
markdown └→ Runner (Playwright CLI) → interactive local testing
| Command | Description |
|---|---|
/setup-profiles | Create or refresh Playwright auth profiles |
/run-qa [smoke|ux|adversarial|all] | Discover screens, confirm manifest, dispatch QA agents |
Framework support: Route discovery is optimized for Next.js (App Router and Pages Router), with support for React Router, Remix, and SvelteKit. Other frameworks fall back to generic route-pattern matching.
| Skill | Trigger | Description |
|---|---|---|
| desktop-workflow-generator | "generate desktop workflows" | Explores codebase, walks the live app with you, co-authors verifications and edge cases |
| mobile-workflow-generator | "generate mobile workflows" | Mobile viewport (393x852), iOS HIG awareness, UX anti-pattern detection |
| multi-user-workflow-generator | "generate multi-user workflows" | Per-persona browser contexts, sync verifications |
| Skill | Trigger | Description |
|---|---|---|
| desktop-workflow-to-playwright | "convert desktop workflows to playwright" | e2e/desktop/ — Chromium tests, auth setup, CI workflow |
| mobile-workflow-to-playwright | "convert mobile workflows to playwright" | e2e/mobile/ — Chromium + WebKit, UX anti-pattern assertions |
| multi-user-workflow-to-playwright | "convert multi-user workflows to playwright" | e2e/multi-user/ — per-persona auth, multi-context patterns |
| Skill | Trigger | Description |
|---|---|---|
| playwright-runner | "run workflows" | Executes workflow markdown interactively via Playwright CLI |
| Skill | Trigger | Description |
|---|---|---|
| adversarial-audit | "adversarial audit" | Maps economic surface area, generates abuse cases across 7 categories |
| resilience-audit | "resilience audit" | Finds breakage from unexpected user behavior — dead ends, race conditions, interrupted ops |
| keyword-wedge | "keyword wedge" | Cross-references codebase with Search Console, PostHog, and Keyword Planner for SEO footholds |
| trust-builder | "trust builder" | Finds free-value trust-building opportunities before asking for commitment |
| review-learnings | "review learnings" | Synthesizes accumulated QA observations into prioritized plugin improvements |
| submit-learnings | "submit learnings" | Filters and submits QA observations as GitHub issues on the plugin repo |
| Skill | Trigger | Description |
|---|---|---|
| use-profiles | Automatic | Loads saved auth profiles before browser automation |
Autonomous QA agents that navigate the app, inspect screens, and produce structured reports. Each agent records observations to a learnings ledger after its session.