Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Build accessible, composable UI with @base-ui/react primitives and project styling tokens. Use when creating or refining unstyled primitive components and behavior-heavy UI such as overlays, menus, popovers, dialogs, and form controls, or defining custom component APIs. Prefer the moai-library-shadcn skill when copied-in shadcn components are the primary implementation path.
.claude/skills/asymmetric-al-base-ui/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | -18% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -58% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 5% | 0% |
| case-18 | ✗→✓ | ▲ Improved | -7% | 0% |
| case-01 | ✓→✓ | = Same ✓ | 1% | 0% |
Name: base-ui Purpose: Build accessible, composable UI with @base-ui/react primitives and project styling tokens. Use this when creating or refining unstyled primitive components and behavior-heavy UI.
Applies when: Building overlays, menus, form controls, or custom component APIs with Base UI. Do not use when: You are using copied-in shadcn components as the primary implementation path (use moai-library-shadcn first).
@base-ui/react is present in the target workspace package.json.cn, tokens, Tailwind).tsxtype FieldShellProps = { label: string; error?: string; children: React.ReactNode; }; export function FieldShell({ label, error, children }: FieldShellProps) { return ( <div className="grid gap-1.5"> <span className="text-sm font-medium">{label}</span> {children} {error ? <p className="text-sm text-destructive">{error}</p> : null} </div> ); }
txtWhen docs examples rely on Tailwind v4-only utilities, replace them with equivalent v3-safe utilities before commit.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 16,976 | 13,263 | -22% | 1 | 1 | 0% | 3,173 | 3,196 | +1% | 0 | 0 | — |
case-02 | pass→pass | 14,468 | 12,900 | -11% | 1 | 1 | 0% | 2,479 | 3,361 | +36% | 0 | 0 | — |
case-03 | pass→pass | 19,954 | 19,202 | -4% | 1 | 1 | 0% | 3,324 | 4,089 | +23% | 0 | 0 | — |
case-04 | pass→pass | 14,215 | 10,361 | -27% | 1 | 1 | 0% | 2,249 | 2,579 | +15% | 0 | 0 | — |
case-05 | pass→pass | 13,394 | 5,958 | -56% | 1 | 1 | 0% | 1,942 | 1,601 | -18% | 0 | 0 | — |
case-06 | pass→pass | 22,017 | 15,912 | -28% | 1 | 1 | 0% | 3,838 | 3,451 | -10% | 0 | 0 | — |
case-07 | fail→pass | 10,807 | 3,399 | -69% | 1 | 1 | 0% | 1,557 | 1,271 | -18% | 0 | 0 | — |
case-08 | fail→pass | 13,700 | 1,364 | -90% | 1 | 1 | 0% | 2,172 | 908 | -58% | 0 | 0 | — |
case-09 | pass→pass | 16,475 | 7,993 | -51% | 1 | 1 | 0% | 2,509 | 2,026 | -19% | 0 | 0 | — |
case-10 | pass→pass | 9,963 | 6,010 | -40% | 1 | 1 | 0% | 1,555 | 1,752 | +13% | 0 | 0 | — |
case-11 | pass→pass | 16,116 | 10,810 | -33% | 1 | 1 | 0% | 2,576 | 2,430 | -6% | 0 | 0 | — |
case-12 | pass→pass | 10,986 | 3,400 | -69% | 1 | 1 | 0% | 1,797 | 1,270 | -29% | 0 | 0 | — |
case-13 | pass→pass | 10,210 | 9,806 | -4% | 1 | 1 | 0% | 1,844 | 2,489 | +35% | 0 | 0 | — |
case-14 | pass→pass | 13,383 | 10,938 | -18% | 1 | 1 | 0% | 2,093 | 2,856 | +36% | 0 | 0 | — |
case-15 | fail→pass | 12,369 | 6,699 | -46% | 1 | 1 | 0% | 1,769 | 1,852 | +5% | 0 | 0 | — |
case-16 | pass→pass | 10,863 | 6,863 | -37% | 1 | 1 | 0% | 1,828 | 1,835 | +0% | 0 | 0 | — |
case-17 | pass→pass | 15,910 | 6,555 | -59% | 1 | 1 | 0% | 2,368 | 1,750 | -26% | 0 | 0 | — |
case-18 | fail→pass | 15,129 | 9,290 | -39% | 1 | 1 | 0% | 2,355 | 2,192 | -7% | 0 | 0 | — |
case-19 | pass→pass | 13,344 | 11,622 | -13% | 1 | 1 | 0% | 2,090 | 2,702 | +29% | 0 | 0 | — |
case-20 | pass→pass | 11,276 | 12,017 | +7% | 1 | 1 | 0% | 1,659 | 2,849 | +72% | 0 | 0 | — |
case-21 | pass→pass | 14,622 | 8,368 | -43% | 1 | 1 | 0% | 2,267 | 2,052 | -9% | 0 | 0 | — |
case-22 | pass→pass | 8,423 | 7,370 | -13% | 1 | 1 | 0% | 1,524 | 2,132 | +40% | 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 +18 percentage points is the difference between those two pass rates over the 22 comparable cases.
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.