Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Modern, clean UI/UX guidance + review skill. Use when you need actionable UX/UI recommendations, design principles, or a design review checklist for new features or existing systems (web/app). Focus on CRAP (Contrast/Repetition/Alignment/Proximity) plus task-first UX, information architecture, feedback & system status, consistency, affordances, error prevention/recovery, and cognitive load. Enforce a modern minimal style (clean, spacious, typography-led), reduce unnecessary copy, forbid emoji as
.claude/skills/davepoon-oiloil-ui-ux-guide/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 6% | 0% |
| case-05 | ✓→✗ | ▼ Worse | -16% | 0% |
| case-16 | ✓→✗ | ▼ Worse | -1% | 0% |
| case-22 | ✓→✗ | ▼ Worse | -12% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | -26% | 0% |
Modern UI/UX guidance and review skill with two modes:
guide: Actionable do/don't rules for modern clean UI/UXreview: Prioritized P0/P1/P2 fix lists with design psychology diagnosisFull skill with references available at: https://github.com/oil-oil/oiloil-ui-ux-guide
Install via: npx skills add oil-oil/oiloil-ui-ux-guide
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 12,241 | 12,830 | +5% | 1 | 1 | 0% | 2,201 | 2,329 | +6% | 0 | 0 | — |
case-02 | pass→pass | 11,843 | 8,244 | -30% | 1 | 1 | 0% | 1,975 | 1,457 | -26% | 0 | 0 | — |
case-03 | pass→pass | 8,816 | 9,591 | +9% | 1 | 1 | 0% | 1,628 | 1,642 | +1% | 0 | 0 | — |
case-04 | pass→pass | 13,123 | 11,098 | -15% | 1 | 1 | 0% | 2,270 | 1,997 | -12% | 0 | 0 | — |
case-05 | pass→fail | 10,818 | 8,524 | -21% | 1 | 1 | 0% | 1,824 | 1,526 | -16% | 0 | 0 | — |
case-06 | pass→pass | 13,338 | 13,883 | +4% | 1 | 1 | 0% | 2,623 | 2,653 | +1% | 0 | 0 | — |
case-07 | pass→pass | 12,708 | 11,590 | -9% | 1 | 1 | 0% | 2,201 | 1,990 | -10% | 0 | 0 | — |
case-08 | pass→pass | 9,671 | 6,938 | -28% | 1 | 1 | 0% | 1,634 | 1,199 | -27% | 0 | 0 | — |
case-09 | pass→pass | 11,602 | 9,942 | -14% | 1 | 1 | 0% | 2,077 | 1,663 | -20% | 0 | 0 | — |
case-10 | fail→fail | 12,516 | 9,104 | -27% | 1 | 1 | 0% | 2,161 | 1,688 | -22% | 0 | 0 | — |
case-11 | pass→pass | 12,940 | 13,369 | +3% | 1 | 1 | 0% | 2,108 | 2,258 | +7% | 0 | 0 | — |
case-12 | pass→pass | 11,820 | 10,626 | -10% | 1 | 1 | 0% | 2,030 | 1,912 | -6% | 0 | 0 | — |
case-13 | pass→pass | 14,156 | 12,361 | -13% | 1 | 1 | 0% | 2,448 | 2,503 | +2% | 0 | 0 | — |
case-14 | fail→fail | 10,899 | 9,600 | -12% | 1 | 1 | 0% | 1,865 | 1,632 | -12% | 0 | 0 | — |
case-15 | pass→pass | 12,750 | 13,308 | +4% | 1 | 1 | 0% | 2,503 | 2,535 | +1% | 0 | 0 | — |
case-16 | pass→fail | 14,364 | 13,669 | -5% | 1 | 1 | 0% | 2,303 | 2,276 | -1% | 0 | 0 | — |
case-17 | pass→pass | 9,485 | 6,734 | -29% | 1 | 1 | 0% | 1,670 | 1,222 | -27% | 0 | 0 | — |
case-18 | fail→fail | 11,440 | 10,664 | -7% | 1 | 1 | 0% | 1,978 | 1,847 | -7% | 0 | 0 | — |
case-19 | pass→pass | 14,392 | 11,460 | -20% | 1 | 1 | 0% | 2,486 | 2,212 | -11% | 0 | 0 | — |
case-20 | pass→pass | 5,428 | 4,657 | -14% | 1 | 1 | 0% | 988 | 916 | -7% | 0 | 0 | — |
case-21 | pass→pass | 13,094 | 11,865 | -9% | 1 | 1 | 0% | 2,251 | 2,193 | -3% | 0 | 0 | — |
case-22 | pass→fail | 7,851 | 6,423 | -18% | 1 | 1 | 0% | 1,546 | 1,360 | -12% | 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 -50 percentage points is the difference between those two pass rates over the 22 comparable cases. 3 cases got worse with the skill loaded, and they are included in that figure.
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.