Install any skill in seconds. Free to start, no credit card required.
Get Started Free →フロントエンド設計の方向性、美的原則、および一貫した設計言語実装。
.claude/skills/affaan-m-frontend-design-direction/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 17% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 49% | 0% |
| case-03 | ✓→✗ | ▼ Worse | 73% | 0% |
| case-06 | ✓→✓ | = Same ✓ | 27% | 0% |
| case-07 | ✓→✓ | = Same ✓ | 46% | 0% |
Use this skill when the work is not just making UI function, but making it feel purposeful, polished, and appropriate to the product domain.
Source: salvaged from stale community PR #1659 by linus707.
Note: ECC intentionally does not rebundle the canonical Anthropic frontend-design skill. Install that from anthropics/skills when you want the official upstream skill. This skill is the ECC-specific design-direction salvage of the useful local guidance from #1659.
less generic.
and interaction choices.
the audience.
Before coding, choose a specific direction:
maximal, minimal, dense, calm, or another explicit direction.
existing design system.
Match the direction to the domain. A SaaS operations tool should usually be dense, quiet, and scannable. A portfolio, launch page, game, or editorial piece can be more expressive. Do not force a landing-page composition onto a tool that needs repeated daily use.
explicitly asks for marketing copy.
before introducing a new visual system.
products, places, people, gameplay, charts, or inspectable media.
coherent across states.
sizes, stable toolbars, and fixed-format controls should not shift when labels or hover states appear.
clarify state over decorative animation.
cleanly rather than overflowing.
blobs, oversized cards, vague hero copy, or stock-like atmospheric media.
restraint.
marketing sections.
itself.
for themselves.
controls, tiles, and counters.
clear reason to depart.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 32,187 | 31,365 | -3% | 1 | 1 | 0% | 6,203 | 7,092 | +14% | 0 | 0 | — |
case-02 | fail→fail | 31,363 | 29,819 | -5% | 1 | 1 | 0% | 6,196 | 7,085 | +14% | 0 | 0 | — |
case-03 | pass→fail | 15,424 | 18,655 | +21% | 1 | 1 | 0% | 2,416 | 4,176 | +73% | 0 | 0 | — |
case-04 | fail→pass | 16,923 | 16,438 | -3% | 1 | 1 | 0% | 3,541 | 4,155 | +17% | 0 | 0 | — |
case-05 | fail→pass | 19,225 | 23,746 | +24% | 1 | 1 | 0% | 3,792 | 5,646 | +49% | 0 | 0 | — |
case-06 | pass→pass | 14,482 | 12,499 | -14% | 1 | 1 | 0% | 2,355 | 2,995 | +27% | 0 | 0 | — |
case-07 | pass→pass | 10,998 | 11,778 | +7% | 1 | 1 | 0% | 1,807 | 2,645 | +46% | 0 | 0 | — |
case-08 | pass→pass | 17,103 | 12,989 | -24% | 1 | 1 | 0% | 2,752 | 2,891 | +5% | 0 | 0 | — |
case-09 | pass→pass | 14,158 | 12,094 | -15% | 1 | 1 | 0% | 1,988 | 2,729 | +37% | 0 | 0 | — |
case-10 | pass→pass | 13,252 | 12,272 | -7% | 1 | 1 | 0% | 1,974 | 2,592 | +31% | 0 | 0 | — |
case-11 | pass→pass | 10,939 | 11,400 | +4% | 1 | 1 | 0% | 1,607 | 2,460 | +53% | 0 | 0 | — |
case-12 | pass→pass | 16,532 | 11,506 | -30% | 1 | 1 | 0% | 2,207 | 2,609 | +18% | 0 | 0 | — |
case-13 | pass→pass | 12,367 | 17,353 | +40% | 1 | 1 | 0% | 1,810 | 2,522 | +39% | 0 | 0 | — |
case-14 | pass→pass | 12,291 | 11,108 | -10% | 1 | 1 | 0% | 1,865 | 2,430 | +30% | 0 | 0 | — |
case-15 | pass→pass | 10,891 | 11,124 | +2% | 1 | 1 | 0% | 1,753 | 2,854 | +63% | 0 | 0 | — |
case-16 | pass→pass | 10,543 | 6,300 | -40% | 1 | 1 | 0% | 1,583 | 1,796 | +13% | 0 | 0 | — |
case-17 | pass→pass | 10,675 | 7,395 | -31% | 1 | 1 | 0% | 1,544 | 2,041 | +32% | 0 | 0 | — |
case-18 | fail→fail | 19,449 | 30,756 | +58% | 1 | 1 | 0% | 2,844 | 6,284 | +121% | 0 | 0 | — |
case-19 | pass→pass | 9,818 | 10,559 | +8% | 1 | 1 | 0% | 1,801 | 2,812 | +56% | 0 | 0 | — |
case-20 | pass→pass | 5,895 | 7,008 | +19% | 1 | 1 | 0% | 972 | 2,115 | +118% | 0 | 0 | — |
case-21 | pass→pass | 10,299 | 11,698 | +14% | 1 | 1 | 0% | 1,884 | 2,877 | +53% | 0 | 0 | — |
case-22 | pass→pass | 13,133 | 10,937 | -17% | 1 | 1 | 0% | 2,111 | 2,898 | +37% | 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 +5 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is 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.