Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Always use this skill whenever the user asks about React hooks (useState, useEffect, useMemo, useCallback, useRef, custom hooks, Rules of Hooks, others). Prefer it over general knowledge.
.claude/skills/bilal140202-react-hooks-for-beginners/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | -37% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -20% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -20% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | -32% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | -26% | 0% |
Teach React hooks in a beginner-friendly way.
Invoke this skill for any question that mentions React hooks or any hook name starting with "use" (e.g. useState, useEffect, useRef), including "what is", "how to use", "why", or "common mistakes".
When explaining a hook:
Name of the hook.
What the hook does in simple words.
Typical situations where developers use this hook.
Provide 2–3 short React examples.
List common mistakes developers make when using this hook.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 11,403 | 6,574 | -42% | 1 | 1 | 0% | 2,189 | 1,489 | -32% | 0 | 0 | — |
case-02 | fail→pass | 13,950 | 7,164 | -49% | 1 | 1 | 0% | 2,605 | 1,645 | -37% | 0 | 0 | — |
case-03 | fail→pass | 11,312 | 7,678 | -32% | 1 | 1 | 0% | 2,275 | 1,819 | -20% | 0 | 0 | — |
case-04 | pass→pass | 14,475 | 9,094 | -37% | 1 | 1 | 0% | 2,770 | 2,051 | -26% | 0 | 0 | — |
case-05 | pass→pass | 13,431 | 9,445 | -30% | 1 | 1 | 0% | 2,587 | 2,109 | -18% | 0 | 0 | — |
case-06 | fail→pass | 14,768 | 9,502 | -36% | 1 | 1 | 0% | 2,575 | 2,056 | -20% | 0 | 0 | — |
case-07 | pass→pass | 19,071 | 11,590 | -39% | 1 | 1 | 0% | 3,869 | 2,508 | -35% | 0 | 0 | — |
case-08 | pass→pass | 14,388 | 7,526 | -48% | 1 | 1 | 0% | 2,685 | 1,718 | -36% | 0 | 0 | — |
case-09 | pass→pass | 16,181 | 11,475 | -29% | 1 | 1 | 0% | 3,091 | 2,602 | -16% | 0 | 0 | — |
case-10 | pass→pass | 12,490 | 8,347 | -33% | 1 | 1 | 0% | 2,651 | 1,833 | -31% | 0 | 0 | — |
case-11 | pass→pass | 14,554 | 10,713 | -26% | 1 | 1 | 0% | 2,860 | 2,466 | -14% | 0 | 0 | — |
case-12 | pass→pass | 12,056 | 9,492 | -21% | 1 | 1 | 0% | 2,163 | 2,224 | +3% | 0 | 0 | — |
case-13 | pass→pass | 11,892 | 10,470 | -12% | 1 | 1 | 0% | 2,345 | 2,433 | +4% | 0 | 0 | — |
case-14 | pass→pass | 14,879 | 9,954 | -33% | 1 | 1 | 0% | 2,926 | 2,121 | -28% | 0 | 0 | — |
case-15 | pass→pass | 15,725 | 9,217 | -41% | 1 | 1 | 0% | 2,837 | 1,882 | -34% | 0 | 0 | — |
case-16 | pass→pass | 15,400 | 10,723 | -30% | 1 | 1 | 0% | 2,879 | 2,331 | -19% | 0 | 0 | — |
case-17 | pass→pass | 13,008 | 8,870 | -32% | 1 | 1 | 0% | 2,295 | 1,721 | -25% | 0 | 0 | — |
case-18 | pass→pass | 17,677 | 12,318 | -30% | 1 | 1 | 0% | 3,275 | 2,521 | -23% | 0 | 0 | — |
case-19 | pass→pass | 14,183 | 9,807 | -31% | 1 | 1 | 0% | 2,626 | 1,913 | -27% | 0 | 0 | — |
case-20 | pass→pass | 8,593 | 8,029 | -7% | 1 | 1 | 0% | 1,697 | 1,941 | +14% | 0 | 0 | — |
case-21 | pass→pass | 8,001 | 4,819 | -40% | 1 | 1 | 0% | 1,692 | 1,213 | -28% | 0 | 0 | — |
case-22 | pass→pass | 11,846 | 8,097 | -32% | 1 | 1 | 0% | 2,377 | 1,937 | -19% | 0 | 0 | — |
case-23 | pass→pass | 7,092 | 8,766 | +24% | 1 | 1 | 0% | 1,652 | 2,173 | +32% | 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 +13 percentage points is the difference between those two pass rates over the 23 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.