browser-test
Browser Test v3.0
Visual verification using Chrome DevTools MCP and Playwright.
Quick Start
/browser-test http://localhost:3000
/browser-test http://localhost:3000 --a11y
/browser-test http://localhost:3000 --perf
Workflow
1. Start Dev Server (if needed)
# Auto-detect package manager and start
npm run dev & # or yarn dev, bun dev
# Wait for server to be ready
2. Navigate and Screenshot
Using Playwright MCP or Chrome DevTools:
- Navigate to URL
- Take full-page screenshot
- Save to
.claude/quality-results/screenshots/
3. Console Errors Check
- Capture all console messages (errors, warnings)
- Flag any errors as BLOCKING
- Flag warnings as ADVISORY
4. Lighthouse Audit
Run via Chrome DevTools MCP:
- Performance score (target: > 90)
- Accessibility score (target: > 90)
- Best Practices score (target: > 90)
- SEO score (target: > 90)
5. Network Check
- Verify no failed requests (4xx, 5xx)
- Check for oversized assets (> 500KB)
- Verify no mixed content (HTTP on HTTPS)
Output Format
Browser Test Report — http://localhost:3000
=============================================
Screenshot: .claude/quality-results/screenshots/2026-04-04-home.png
Console: 0 errors, 2 warnings
Lighthouse: Perf 94 | A11y 98 | BP 100 | SEO 92
Network: 23 requests, 0 failed, largest: 145KB
---------------------------------------------
RESULT: PASS (0 blocking, 2 advisory)
Integration Points
| Skill | How |
|---|---|
| /gates Stage 5 | BROWSER validation (advisory for frontend) |
| Orchestrator Step 7 | Include browser testing for frontend tasks |
| ralph-frontend | Invoke via skill, not via MCP directly |
Accessibility Audit Mode (--a11y) (v3.0 — Item 13)
/browser-test http://localhost:3000 --a11y
Runs axe-core via Lighthouse accessibility audit:
- WCAG 2.1 AA compliance check
- Color contrast violations
- Missing alt text, labels, ARIA roles
- Keyboard navigation traps
- Focus management issues
Output blocked in ralph-frontend quality gate when score < 90.
Reference: docs/reference/accessibility-checklist.md
Visual Regression Mode (v3.0 — Item 14)
Screenshots saved to .claude/quality-results/screenshots/ with timestamp.
For visual comparison between runs:
# Compare current vs previous screenshot
mcp__zai-mcp-server__ui_diff_check(before_image, after_image)
Screenshots are gitignored (.claude/quality-results in .gitignore).
Anti-Rationalization
| Excuse | Rebuttal |
|---|---|
| "It looks fine in the code" | Visual bugs are invisible in code review. Test in browser. |
| "Lighthouse is too strict" | Lighthouse catches real user experience issues. |
| "Console warnings don't matter" | Warnings become errors in strict mode and for users. |
| "Accessibility is for later" | WCAG 2.1 AA is mandatory, not optional. |
More from alfredolopez80/multi-agent-ralph-loop
stop-slop
A skill for removing AI-generated writing patterns ('slop') from prose. Eliminates telltale signs of AI writing like filler phrases, excessive hedging, overly formal language, and mechanical sentence structures. Use when: writing content that should sound human and natural, editing AI-generated drafts, cleaning up prose for publication, or any content that needs to sound authentic rather than AI-generated. Triggers: 'stop-slop', 'remove AI tells', 'clean up prose', 'make it sound human', 'edit AI writing'.
10iterate
Ralph Loop pattern with swarm mode: iterative execution until VERIFIED_DONE with multi-agent coordination. Use when: (1) iterative refinement needed, (2) quality gates must pass, (3) automated validation required. Triggers: /iterate, 'iterate until done', 'keep trying', 'fix until passing', 'loop until done'.
2gemini-cli
|
2minimax
Custom skill for minimax
1clarify
Intensive requirement clarification using structured AskUserQuestion workflow. Gathers MUST_HAVE (blocking) and NICE_TO_HAVE (optional) information before implementation. Use when: (1) starting new feature implementation, (2) requirements are ambiguous, (3) multiple approaches possible, (4) before writing any code. Triggers: /clarify, 'clarify requirements', 'ask questions', 'gather requirements'.
1security
Security audit with Codex + MiniMax second opinion. Integrates ralph-security agent (6 quality pillars, OWASP A01-A10). Uses LSP for code navigation during analysis. Use when: (1) /security is invoked, (2) task relates to security functionality.
1