Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Apple Human Interface Guidelines for platform-specific design.
.claude/skills/sickn33-hig-platforms/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 30% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 55% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 32% | 0% |
| case-18 | ✗→✓ | ▲ Improved | 54% | 0% |
| case-22 | ✗→✓ | ▲ Improved | 48% | 0% |
Check for .claude/apple-design-context.md before asking questions. Use existing context and only ask for information not already covered.
| Reference | Topic | Key content | |---|---|---| | designing-for-ios.md | iOS | Touch, tab bars, navigation stacks, gestures, screen sizes, safe areas | | designing-for-ipados.md | iPadOS | Multitasking, sidebars, pointer, keyboard, Apple Pencil, Stage Manager | | designing-for-macos.md | macOS | Menu bars, toolbars, window management, keyboard shortcuts, dense layouts, Dock | | designing-for-tvos.md | tvOS | Focus engine, Siri Remote, lean-back experience, content-forward, parallax | | designing-for-visionos.md | visionOS | Spatial computing, windows/volumes/spaces, eye tracking, hand gestures, depth | | designing-for-watchos.md | watchOS | Glanceable UI, Digital Crown, complications, notifications, haptics | | designing-for-games.md | Games | Controllers, immersive experiences, platform-specific conventions, accessibility |
Built by Raintree Technology · More developer tools
This skill is applicable to execute the workflow or actions described in the overview.
User request:
> Use @hig-platforms for this task: Apple Human Interface Guidelines for platform-specific design.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 36,518 | 24,953 | -32% | 1 | 1 | 0% | 6,210 | 4,694 | -24% | 0 | 0 | — |
case-02 | fail→pass | 18,670 | 15,501 | -17% | 1 | 1 | 0% | 2,833 | 3,692 | +30% | 0 | 0 | — |
case-03 | fail→fail | 16,715 | 21,256 | +27% | 1 | 1 | 0% | 2,487 | 3,988 | +60% | 0 | 0 | — |
case-04 | fail→fail | 19,590 | 21,525 | +10% | 1 | 1 | 0% | 2,822 | 4,561 | +62% | 0 | 0 | — |
case-05 | fail→fail | 14,396 | 18,378 | +28% | 1 | 1 | 0% | 2,019 | 3,827 | +90% | 0 | 0 | — |
case-06 | fail→fail | 16,679 | 16,585 | -1% | 1 | 1 | 0% | 2,417 | 3,483 | +44% | 0 | 0 | — |
case-07 | fail→fail | 13,490 | 17,914 | +33% | 1 | 1 | 0% | 2,102 | 3,909 | +86% | 0 | 0 | — |
case-08 | fail→fail | 14,678 | 14,071 | -4% | 1 | 1 | 0% | 2,116 | 3,179 | +50% | 0 | 0 | — |
case-09 | pass→pass | 15,838 | 17,872 | +13% | 1 | 1 | 0% | 2,394 | 4,158 | +74% | 0 | 0 | — |
case-10 | fail→fail | 18,640 | 17,211 | -8% | 1 | 1 | 0% | 2,553 | 3,745 | +47% | 0 | 0 | — |
case-11 | pass→pass | 15,617 | 17,242 | +10% | 1 | 1 | 0% | 2,365 | 3,782 | +60% | 0 | 0 | — |
case-12 | pass→pass | 14,829 | 14,625 | -1% | 1 | 1 | 0% | 2,142 | 3,184 | +49% | 0 | 0 | — |
case-13 | fail→fail | 14,326 | 17,693 | +24% | 1 | 1 | 0% | 2,158 | 3,920 | +82% | 0 | 0 | — |
case-14 | fail→pass | 17,049 | 18,313 | +7% | 1 | 1 | 0% | 2,544 | 3,953 | +55% | 0 | 0 | — |
case-15 | pass→pass | 17,794 | 17,937 | +1% | 1 | 1 | 0% | 2,865 | 4,191 | +46% | 0 | 0 | — |
case-16 | pass→pass | 16,862 | 15,487 | -8% | 1 | 1 | 0% | 2,545 | 3,769 | +48% | 0 | 0 | — |
case-17 | fail→pass | 18,506 | 16,728 | -10% | 1 | 1 | 0% | 2,699 | 3,566 | +32% | 0 | 0 | — |
case-18 | fail→pass | 17,311 | 17,833 | +3% | 1 | 1 | 0% | 2,568 | 3,958 | +54% | 0 | 0 | — |
case-19 | pass→pass | 11,389 | 10,555 | -7% | 1 | 1 | 0% | 2,180 | 3,309 | +52% | 0 | 0 | — |
case-20 | pass→pass | 15,270 | 16,324 | +7% | 1 | 1 | 0% | 2,991 | 4,031 | +35% | 0 | 0 | — |
case-21 | pass→pass | 11,303 | 14,342 | +27% | 1 | 1 | 0% | 2,113 | 3,863 | +83% | 0 | 0 | — |
case-22 | fail→pass | 14,311 | 14,674 | +3% | 1 | 1 | 0% | 2,418 | 3,590 | +48% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +23 percentage points is the difference between those two pass rates over the 22 comparable cases.
The publisher has shipped newer versions since this run, so these numbers describe v1, not the version currently listed.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.