Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generate a new UI component that follows StyleSeed Toss conventions for structure, tokens, accessibility, and component ergonomics.
.claude/skills/lingxling-ui-component/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-17 | ✗→✓ | ▲ Improved | -3% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 35% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -19% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 7% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -23% | 0% |
Part of StyleSeed, this skill generates components that respect the Toss seed's design language instead of improvising ad hoc markup and styling. It emphasizes semantic tokens, predictable typing, reusable variants, and mobile-friendly accessibility defaults.
Before generating code, inspect the seed's source of truth:
CLAUDE.md for conventionscss/theme.css for semantic tokenscomponents/ui/If the user already has a better local example, follow the local codebase over a generic template.
Place the output where it belongs:
src/components/ui/ for primitives and low-level building blockssrc/components/patterns/ for composed sections or multi-part patternsDo not create a new primitive if an existing one can be extended safely.
Use these defaults unless the host project strongly disagrees:
const componentReact.ComponentProps<> or equivalent native prop typingclassName passthrough supportcn() or the project's standard class mergerdata-slot for component identificationDo not hardcode visual values if the design system has a token for them.
Preferred examples:
bg-cardtext-foregroundtext-muted-foregroundborder-bordershadow-[var(--shadow-card)]aria-* attributes where appropriateProvide:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-17 | fail→pass | 24,739 | 21,356 | -14% | 1 | 1 | 0% | 5,318 | 5,138 | -3% | 0 | 0 | — |
case-05 | pass→pass | 12,158 | 17,718 | +46% | 1 | 1 | 0% | 2,326 | 4,439 | +91% | 0 | 0 | — |
case-01 | fail→pass | 14,424 | 18,878 | +31% | 1 | 1 | 0% | 3,098 | 4,169 | +35% | 0 | 0 | — |
case-02 | fail→pass | 51,817 | 58,055 | +12% | 1 | 1 | 0% | 7,799 | 6,309 | -19% | 0 | 0 | — |
case-03 | fail→pass | 24,566 | 16,676 | -32% | 1 | 1 | 0% | 3,847 | 4,105 | +7% | 0 | 0 | — |
case-04 | pass→fail | 11,381 | 10,841 | -5% | 1 | 1 | 0% | 1,690 | 2,944 | +74% | 0 | 0 | — |
case-06 | pass→pass | 12,663 | 11,628 | -8% | 1 | 1 | 0% | 1,946 | 2,965 | +52% | 0 | 0 | — |
case-07 | fail→fail | 34,493 | 24,639 | -29% | 1 | 1 | 0% | 5,397 | 4,672 | -13% | 0 | 0 | — |
case-08 | fail→fail | 22,616 | 16,732 | -26% | 1 | 1 | 0% | 3,518 | 4,487 | +28% | 0 | 0 | — |
case-09 | fail→fail | 26,272 | 22,999 | -12% | 1 | 1 | 0% | 4,115 | 5,058 | +23% | 0 | 0 | — |
case-10 | fail→pass | 37,114 | 22,957 | -38% | 1 | 1 | 0% | 5,880 | 4,503 | -23% | 0 | 0 | — |
case-11 | fail→fail | 22,403 | 29,995 | +34% | 1 | 1 | 0% | 3,291 | 5,656 | +72% | 0 | 0 | — |
case-12 | fail→fail | 28,374 | 25,335 | -11% | 1 | 1 | 0% | 6,925 | 6,574 | -5% | 0 | 0 | — |
case-13 | fail→fail | 32,484 | 24,271 | -25% | 1 | 1 | 0% | 5,138 | 4,434 | -14% | 0 | 0 | — |
case-14 | fail→fail | 23,745 | 22,595 | -5% | 1 | 1 | 0% | 4,894 | 4,239 | -13% | 0 | 0 | — |
case-15 | fail→fail | 38,239 | 28,921 | -24% | 1 | 1 | 0% | 6,147 | 5,081 | -17% | 0 | 0 | — |
case-16 | fail→fail | 23,253 | 13,182 | -43% | 1 | 1 | 0% | 3,466 | 3,242 | -6% | 0 | 0 | — |
case-18 | fail→fail | 26,107 | 21,674 | -17% | 1 | 1 | 0% | 5,615 | 4,143 | -26% | 0 | 0 | — |
case-19 | fail→pass | 31,869 | 17,324 | -46% | 1 | 1 | 0% | 4,993 | 4,286 | -14% | 0 | 0 | — |
case-20 | fail→fail | 29,480 | 34,800 | +18% | 1 | 1 | 0% | 4,572 | 5,718 | +25% | 0 | 0 | — |
case-21 | fail→fail | 12,129 | 36,111 | +198% | 1 | 1 | 0% | 2,289 | 2,738 | +20% | 0 | 0 | — |
case-22 | fail→fail | 29,827 | 26,222 | -12% | 1 | 1 | 0% | 4,736 | 4,685 | -1% | 0 | 0 | — |
case-23 | fail→pass | 51,256 | 24,066 | -53% | 1 | 1 | 0% | 5,306 | 5,530 | +4% | 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. 23 cases were attempted. The headline lift of +26 percentage points is the difference between those two pass rates over the 23 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.