JPeetz/agent-skills / skills/frontend/ui-verification
ui-verification
VerifiedUse this skill when asked to verify UI in the browser, reproduce a UI finding, re-measure after a fix, or try to break a feature with browser probes.
| File | Kind | Size |
|---|---|---|
| CHANGELOG.md | markdown | 2.5 KB |
| LICENSE | other | 1.1 KB |
| SKILL.md | skill md | 24.3 KB |
| evals/evals.json | config | 3.2 KB |
| references/exploratory-test.md | markdown | 4.8 KB |
| references/visual-review.md | markdown | 5.2 KB |
CHANGELOG.md
View on GitHub# CHANGELOG — ui-verification
## v0.1.0 — 2026-11-14
### Added
- Initial release of the UI Verification skill.
- IS / IS NOT contract: owns the browser session; does not read source to find
defects, build UI, author durable test suites, or issue ship verdicts.
- Nine-probe catalogue with inline playbooks (no separate probe files, since
this repo ships no browser test infrastructure):
- `axe-scan` — axe-core violations per route and theme, computed contrast.
- `target-size` — bounding box and effective hit area for every visible
interactive element, including pseudo-element expansion.
- `focus-walk` — scripted Tab traversal, focus-ring pixel delta, dialog trap
and focus restoration.
- `layout-shift` — attributed `layout-shift` entries with the data response
held open.
- `viewport-stress` — horizontal overflow and clipped text from 320px up,
plus tripled-string passes.
- `failure-injection` — 500, `[]`, malformed, and aborted responses, with
retry call counting.
- `theme-locale-matrix` — one route across viewport, theme, pseudo-locale,
and direction.
- `console-network` — console errors, page errors, failed requests, hydration
mismatch warnings.
- `web-vitals` — LCP with attributed element, CLS, INP on a scripted
interaction.
- Six-step workflow: session, probe selection, execution, decision
(`reproduced` / `not-reproduced` / `unknown`), clearing re-run with
`clearedBy`, report.
- Feature-to-probe minimum sets for forms, lists, modals, dashboards, checkout.
- Honesty rules: skipped is not passed; screenshots are evidence not verdicts;
the app is the subject, not the harness; numbers carry units and conditions.
- Cross-probe gotchas: dark-mode emulation misses attribute theming, dev-mode
perf numbers, unsettled animations, leaked browser state, late console
listeners.
- `references/exploratory-test.md` — fresh-context exploratory pass against a
test environment.
- `references/visual-review.md` — matched before/after screenshot comparison
and suspicious-duplicate detection.
- Eval scenarios covering triggering, probe selection, and near-miss negatives.
### Why
Static UI audits infer what a user will see. Two classes of defect survive that
inference every time: the control that looks too small but has a pseudo-element
hit area, and the action that looks wired but goes nowhere. Both are one browser
measurement away from being settled. This skill supplies the measurement, the
evidence shape, and the honesty rules that keep a skipped probe from being read
as a pass.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| brainstormingobra/superpowersYou MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation). | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| brand-guidelinesanthropics/skillsApplies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| frontend-designanthropics/skillsGuidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| mcp-builderanthropics/skillsGuide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK). | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |