agentleFS
LucaDominiciLucaDominici/arbiter/.claude/skills/visual-verification/SKILL.md

visual-verification

Skill0 starsChanged 55 days agoApache-2.0View on GitHub

Automated visual regression check via Playwright — 5 DOM checks × 3 viewports (375/768/1280). Produces pass/fail grid + screenshot paths under .arbiter/evidence/visual/.

  • Installs packages

5 other projects use a version of this.

Adopted by

Most recently taken up by

The file

Files
SKILL.md

Visual Verification Skill

Invoke when a UI change needs visual validation before marking a task complete.

Usage

node scripts/visual-verify.mjs --url http://localhost:3000 --task-id 700

# Graceful skip when Playwright not installed:
node scripts/visual-verify.mjs --skip-if-missing --url http://localhost:3000 --task-id 700

What it checks (5 checks × 3 viewports = 15 datapoints)

#CheckDescription
1page-title<title> element present and non-empty
2no-layout-overflowbody scroll width ≤ viewport width
3primary-cta-visibleAt least one <button> or <a> element visible
4no-console-errorNo console errors on page load
5screenshotFull-page screenshot captured for review

Viewports: 375px (mobile), 768px (tablet), 1280px (desktop).

Output schema

Results written to .arbiter/evidence/visual/<task-id>/<page>.json:

{
  "url": "http://localhost:3000",
  "task_id": "700",
  "timestamp": "2026-05-17T11:00:00Z",
  "viewports": [
    {
      "width": 375,
      "checks": {
        "page-title": { "pass": true },
        "no-layout-overflow": { "pass": true },
        "primary-cta-visible": { "pass": false, "detail": "no visible CTA found" },
        "no-console-error": { "pass": true },
        "screenshot": { "pass": true, "path": ".arbiter/evidence/visual/700/375.png" }
      }
    }
  ],
  "summary": { "total": 15, "passed": 14, "failed": 1 }
}

When Playwright is absent

With --skip-if-missing, the script exits 0 and writes a skip marker:

{
  "skipped": true,
  "reason": "Playwright not installed",
  "install": "npm install -D @playwright/test && npx playwright install chromium"
}

Without --skip-if-missing, it prints install instructions and exits 1.

Prerequisites

npm install -D @playwright/test
npx playwright install chromium

Task completion gate

A task with UI changes is complete only when ALL 15 datapoints pass (or the task explicitly waives specific checks with documented justification in the PR description).

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.