Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Cursor rules for Next.js development with Tailwind CSS and SEO optimization.
.claude/skills/amariahak-cursor-rules-for-next/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-13 | ✗→✓ | ▲ Improved | -6% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 6% | 0% |
| case-01 | ✓→✓ | = Same ✓ | -5% | 0% |
| case-03 | ✓→✓ | = Same ✓ | 3% | 0% |
| case-04 | ✓→✓ | = Same ✓ | 56% | 0% |
Cursor rules for Next.js development with React 19, Vercel AI, and Tailwind CSS integration.
Synced from https://github.com/PatrickJS/awesome-cursorrules/tree/main/rules/nextjs15-react19-vercelai-tailwind-cursorrules-prompt-file.mdc.
You are an expert senior software engineer specializing in modern web development, with deep expertise in TypeScript, React 19, Next.js 15 (App Router), Vercel AI SDK, Shadcn UI, Radix UI, and Tailwind CSS. You are thoughtful, precise, and focus on delivering high-quality, maintainable solutions.
Before responding to any request, follow these steps:
satisfies operator for type validationuseActionState instead of deprecated useFormStateuseFormStatus with new properties (data, method, action)typescript// Always use async versions of runtime APIs const cookieStore = await cookies() const headersList = await headers() const { isEnabled } = await draftMode() // Handle async params in layouts/pages const params = await props.params const searchParams = await props.searchParams
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 20,502 | 15,530 | -24% | 1 | 1 | 0% | 4,626 | 4,380 | -5% | 0 | 0 | — |
case-02 | fail→fail | 21,318 | 17,486 | -18% | 1 | 1 | 0% | 4,860 | 4,816 | -1% | 0 | 0 | — |
case-03 | pass→pass | 8,285 | 5,832 | -30% | 1 | 1 | 0% | 1,694 | 1,741 | +3% | 0 | 0 | — |
case-04 | pass→pass | 26,814 | 19,027 | -29% | 1 | 1 | 0% | 3,099 | 4,841 | +56% | 0 | 0 | — |
case-05 | pass→pass | 24,251 | 7,098 | -71% | 1 | 1 | 0% | 3,119 | 2,067 | -34% | 0 | 0 | — |
case-06 | pass→pass | 13,467 | 7,679 | -43% | 1 | 1 | 0% | 2,719 | 2,240 | -18% | 0 | 0 | — |
case-07 | pass→pass | 14,307 | 9,523 | -33% | 1 | 1 | 0% | 2,621 | 2,596 | -1% | 0 | 0 | — |
case-08 | pass→pass | 11,501 | 12,422 | +8% | 1 | 1 | 0% | 2,109 | 3,355 | +59% | 0 | 0 | — |
case-09 | pass→pass | 17,194 | 11,222 | -35% | 1 | 1 | 0% | 3,390 | 2,970 | -12% | 0 | 0 | — |
case-10 | pass→pass | 18,495 | 18,691 | +1% | 1 | 1 | 0% | 4,292 | 4,723 | +10% | 0 | 0 | — |
case-11 | pass→pass | 15,357 | 13,522 | -12% | 1 | 1 | 0% | 3,472 | 2,930 | -16% | 0 | 0 | — |
case-12 | pass→pass | 10,626 | 9,418 | -11% | 1 | 1 | 0% | 2,174 | 2,621 | +21% | 0 | 0 | — |
case-13 | fail→pass | 19,509 | 14,365 | -26% | 1 | 1 | 0% | 4,068 | 3,823 | -6% | 0 | 0 | — |
case-14 | pass→pass | 13,626 | 12,044 | -12% | 1 | 1 | 0% | 2,742 | 3,182 | +16% | 0 | 0 | — |
case-15 | pass→pass | 19,012 | 14,762 | -22% | 1 | 1 | 0% | 4,129 | 4,042 | -2% | 0 | 0 | — |
case-16 | pass→pass | 23,930 | 16,336 | -32% | 1 | 1 | 0% | 5,741 | 4,439 | -23% | 0 | 0 | — |
case-17 | fail→pass | 9,661 | 7,268 | -25% | 1 | 1 | 0% | 1,826 | 1,935 | +6% | 0 | 0 | — |
case-18 | pass→pass | 14,815 | 12,294 | -17% | 1 | 1 | 0% | 2,970 | 3,441 | +16% | 0 | 0 | — |
case-19 | pass→pass | 11,551 | 8,459 | -27% | 1 | 1 | 0% | 2,431 | 2,392 | -2% | 0 | 0 | — |
case-20 | pass→pass | 5,473 | 5,045 | -8% | 1 | 1 | 0% | 1,168 | 1,628 | +39% | 0 | 0 | — |
case-21 | pass→pass | 16,793 | 14,704 | -12% | 1 | 1 | 0% | 3,861 | 3,905 | +1% | 0 | 0 | — |
case-22 | pass→pass | 12,012 | 10,657 | -11% | 1 | 1 | 0% | 2,371 | 2,313 | -2% | 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 +9 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.