screenshot-analyze-verification_skill

This skill verifies UI screenshots against strict gates to ensure correct app, complete elements, readability, and layout integrity across modes.

Python

7

GitHub Stars

1

Bundled Files

3 weeks ago

Catalog Refreshed

2 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstart where the catalogue uses aiagentskills.

npx veilstart add skill vladimirbrejcha/ios-ai-skills --skill screenshot-analyze-verification

SKILL.md7.7 KB

Overview

This skill enforces a strict, fail-fast verification workflow for simulator UI screenshots used as proof of correctness. It validates app/page fidelity, required elements, spotlight suppression, legibility, and layout geometry across requested OS and theme coverage. Any hard-gate failure yields NOT VERIFIED and returns actionable findings.

How this skill works

The skill inspects each screenshot against an explicit input checklist: target app/page, scenario/state, required simulators, theme modes, visible and suppressed elements, and interaction expectations. It runs seven exact validation checks (context, elements, spotlight, copy, spacing, arrow correctness, cross-mode parity) and marks each row PASS/FAIL; any FAIL => NOT VERIFIED. Outputs include gate results, prioritized findings with screenshot paths, required fixes, and residual risks if applicable.

When to use it

Validating tutorial hints, callouts, and spotlight overlays before shipping
Cross-OS and light/dark mode parity checks for UI captures
Pre-merge verification of UI screenshots used as design proof
Spot-checking tap-target visuals and dismissal semantics in simulator captures
Catching regressions like clipping, detached pointers, or suppression leaks

Best practices

Provide complete inputs up front: app/page, scenario label, simulators, themes, expected elements
Rebuild and reinstall the latest binary on each simulator before capture
Capture deterministic filenames including platform, OS, scenario, and theme
Include both light and dark screenshots when theme coverage is required
Annotate required hidden elements and interaction semantics to avoid ambiguity

Example use cases

Approve a Training Plan Card hint overlay across iOS 15/16 in both themes
Verify top toolbar spotlight behavior does not reveal persistent chrome
Detect and file regressions when a bubble pointer detaches after layout clamping
Block a release when instructional text fails contrast thresholds in dark mode
Ensure bubble padding and arrow geometry meet Apple HIG-inspired clarity rules

FAQ

Validation is paused and the skill requests the missing details; do not proceed without complete inputs.

Can borderline contrast pass?

Treat borderline legibility as a fail; aim for WCAG-like ~4.5:1 equivalent for critical text.