Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generate open-ended quiz questions from facts in a specific deck (or all facts). Read-only — does NOT update SM-2 state or review-log. Use when the user says "quiz", "test me", "pratique", "/learn-quiz", or wants to practice without triggering a formal review.
.claude/skills/evolution-foundation-learn-quiz/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-08 | ✗→✓ | ▲ Improved | 39% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -71% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -46% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 16% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | -43% | 0% |
Generates open-ended practice questions from facts in workspace/learning/facts/. This is a read-only exercise — no frontmatter is modified, no log entries are written.
User wants to practice recall without formally reviewing (e.g., "me teste sobre claude-skills", "/learn-quiz", "quiz geral").
Parse the user's request for a deck name:
/learn-quiz {deck-name} → filter facts where deck == deck-name/learn-quiz (no argument) → use all facts in workspace/learning/facts/If no facts are found for the specified deck: > "Nenhum fato encontrado no deck '{deck}'. Verifique o nome do deck ou use /learn-stats para ver os decks ativos."
If workspace/learning/facts/ is empty or doesn't exist: > "Nenhum fato encontrado. Use /learn-capture para adicionar fatos primeiro."
From the matching facts, select up to 5 to quiz on. Prefer:
next_review)For each selected fact, generate one open-ended question derived from the Retrieval Q field. You may rephrase the original question slightly for variety, but the answer must still be the Fact content.
Rules for questions:
For each question (up to 5):
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
🧪 Quiz — {deck} | Pergunta {current}/{total}
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
{question}Wait for the user's answer. Then show:
✅ Resposta de referência:
{Fact content}
{Brief feedback: correct elements the user got right, what they missed — 1-2 sentences}Do NOT score formally. This is practice, not review.
After all questions:
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
🏁 Quiz concluído! {N} perguntas praticadas.
💡 Para revisão formal com SM-2, use /learn-review
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-20 | fail→fail | 6,562 | 4,773 | -27% | 1 | 1 | 0% | 1,179 | 1,317 | +12% | 0 | 0 | — |
case-01 | fail→fail | 8,544 | 4,611 | -46% | 1 | 1 | 0% | 1,313 | 1,122 | -15% | 0 | 0 | — |
case-02 | fail→fail | 9,041 | 2,807 | -69% | 1 | 1 | 0% | 1,217 | 916 | -25% | 0 | 0 | — |
case-03 | fail→fail | 12,758 | 8,488 | -33% | 1 | 1 | 0% | 2,052 | 845 | -59% | 0 | 0 | — |
case-04 | pass→pass | 11,491 | 3,020 | -74% | 1 | 1 | 0% | 2,086 | 1,185 | -43% | 0 | 0 | — |
case-05 | fail→fail | 3,987 | 16,963 | +325% | 1 | 1 | 0% | 134 | 2,802 | +1991% | 0 | 0 | — |
case-06 | fail→fail | 13,668 | 2,760 | -80% | 1 | 1 | 0% | 2,134 | 898 | -58% | 0 | 0 | — |
case-07 | pass→pass | 16,662 | 3,537 | -79% | 1 | 1 | 0% | 2,001 | 1,197 | -40% | 0 | 0 | — |
case-08 | fail→pass | 4,532 | 2,185 | -52% | 1 | 1 | 0% | 708 | 981 | +39% | 0 | 0 | — |
case-09 | fail→pass | 18,085 | 3,208 | -82% | 1 | 1 | 0% | 3,576 | 1,033 | -71% | 0 | 0 | — |
case-10 | fail→fail | 13,399 | 11,547 | -14% | 1 | 1 | 0% | 1,832 | 2,093 | +14% | 0 | 0 | — |
case-11 | fail→pass | 11,517 | 3,713 | -68% | 1 | 1 | 0% | 2,132 | 1,160 | -46% | 0 | 0 | — |
case-12 | fail→fail | 4,555 | 2,701 | -41% | 1 | 1 | 0% | 751 | 977 | +30% | 0 | 0 | — |
case-13 | fail→pass | 4,994 | 2,644 | -47% | 1 | 1 | 0% | 897 | 1,042 | +16% | 0 | 0 | — |
case-14 | pass→pass | 6,119 | 2,218 | -64% | 1 | 1 | 0% | 1,074 | 1,008 | -6% | 0 | 0 | — |
case-15 | fail→fail | 7,621 | 5,372 | -30% | 1 | 1 | 0% | 1,251 | 953 | -24% | 0 | 0 | — |
case-16 | fail→fail | 7,106 | 6,481 | -9% | 1 | 1 | 0% | 1,309 | 942 | -28% | 0 | 0 | — |
case-17 | fail→fail | 3,699 | 3,335 | -10% | 1 | 1 | 0% | 558 | 1,171 | +110% | 0 | 0 | — |
case-18 | fail→fail | 7,162 | 3,441 | -52% | 1 | 1 | 0% | 1,206 | 1,005 | -17% | 0 | 0 | — |
case-19 | fail→fail | 2,932 | 5,660 | +93% | 1 | 1 | 0% | 487 | 1,027 | +111% | 0 | 0 | — |
case-21 | fail→fail | 4,142 | 2,701 | -35% | 1 | 1 | 0% | 555 | 957 | +72% | 0 | 0 | — |
case-22 | pass→pass | 8,986 | 5,353 | -40% | 1 | 1 | 0% | 1,474 | 983 | -33% | 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, and 17 counted toward the lift figure. The other 5 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +18 percentage points is the difference between those two pass rates over the 17 comparable cases. 3 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.