Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →You are a React component architecture expert specializing in scaffolding production-ready, accessible, and performant components. Generate complete component implementations with TypeScript, tests, s
.claude/skills/dokhacgiakhoa-frontend-mobile-development-component-scaffold/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✗→✓ | ▲ Improved | -1% | 0% |
| case-09 | ✓→✗ | ▼ Worse | -6% | 0% |
| case-17 | ✓→✗ | ▼ Worse | -24% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | -3% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | -2% | 0% |
You are a React component architecture expert specializing in scaffolding production-ready, accessible, and performant components. Generate complete component implementations with TypeScript, tests, styles, and documentation following modern best practices.
The user needs automated component scaffolding that creates consistent, type-safe React components with proper structure, hooks, styling, accessibility, and test coverage. Focus on reusable patterns and scalable architecture.
$ARGUMENTS
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 41,151 | 41,862 | +2% | 1 | 1 | 0% | 7,084 | 6,872 | -3% | 0 | 0 | — |
case-02 | fail→fail | 47,555 | 38,590 | -19% | 1 | 1 | 0% | 8,236 | 8,575 | +4% | 0 | 0 | — |
case-03 | pass→pass | 37,044 | 34,291 | -7% | 1 | 1 | 0% | 8,232 | 8,038 | -2% | 0 | 0 | — |
case-04 | fail→fail | 37,121 | 29,466 | -21% | 1 | 1 | 0% | 8,226 | 7,402 | -10% | 0 | 0 | — |
case-05 | pass→pass | 35,490 | 27,053 | -24% | 1 | 1 | 0% | 5,873 | 6,016 | +2% | 0 | 0 | — |
case-06 | fail→pass | 47,621 | 43,836 | -8% | 1 | 1 | 0% | 8,230 | 8,182 | -1% | 0 | 0 | — |
case-07 | fail→fail | 45,707 | 46,866 | +3% | 1 | 1 | 0% | 8,228 | 8,567 | +4% | 0 | 0 | — |
case-08 | fail→fail | 44,748 | 40,915 | -9% | 1 | 1 | 0% | 8,071 | 8,168 | +1% | 0 | 0 | — |
case-09 | pass→fail | 41,826 | 26,003 | -38% | 1 | 1 | 0% | 6,967 | 6,523 | -6% | 0 | 0 | — |
case-10 | pass→pass | 31,364 | 25,866 | -18% | 1 | 1 | 0% | 6,739 | 6,425 | -5% | 0 | 0 | — |
case-11 | pass→pass | 32,307 | 26,502 | -18% | 1 | 1 | 0% | 5,840 | 6,537 | +12% | 0 | 0 | — |
case-12 | pass→pass | 31,906 | 43,542 | +36% | 1 | 1 | 0% | 7,867 | 8,226 | +5% | 0 | 0 | — |
case-13 | pass→pass | 23,037 | 22,641 | -2% | 1 | 1 | 0% | 5,853 | 5,384 | -8% | 0 | 0 | — |
case-14 | pass→pass | 30,703 | 24,269 | -21% | 1 | 1 | 0% | 7,128 | 6,308 | -12% | 0 | 0 | — |
case-15 | fail→fail | 30,232 | 41,717 | +38% | 1 | 1 | 0% | 6,961 | 7,772 | +12% | 0 | 0 | — |
case-16 | pass→pass | 31,910 | 32,889 | +3% | 1 | 1 | 0% | 7,683 | 6,768 | -12% | 0 | 0 | — |
case-17 | pass→fail | 34,360 | 23,867 | -31% | 1 | 1 | 0% | 7,578 | 5,755 | -24% | 0 | 0 | — |
case-18 | pass→pass | 31,343 | 29,291 | -7% | 1 | 1 | 0% | 7,083 | 6,794 | -4% | 0 | 0 | — |
case-19 | pass→pass | 39,063 | 27,191 | -30% | 1 | 1 | 0% | 6,702 | 6,747 | +1% | 0 | 0 | — |
case-20 | pass→pass | 19,125 | 33,248 | +74% | 1 | 1 | 0% | 3,111 | 6,036 | +94% | 0 | 0 | — |
case-21 | pass→pass | 14,216 | 22,288 | +57% | 1 | 1 | 0% | 2,900 | 4,004 | +38% | 0 | 0 | — |
case-22 | pass→pass | 24,233 | 27,831 | +15% | 1 | 1 | 0% | 3,869 | 5,202 | +34% | 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 -33 percentage points is the difference between those two pass rates over the 22 comparable cases. 2 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.