flashcards
/flashcards
Tworzy, testuje i aktualizuje fiszki do aktywnego przypominania oparte na pojęciach z sesji, aby wzmocnić zapamiętywanie.
- Autor
- yugash007 ↗
- Repozytorium
- edu-agent-skills ↗
- Licencja
- MIT
- Ostatnia zmiana repozytorium
- 19.05.2026
Ocena i bezpieczeństwo
- Pliki w paczce
- 3
- Skrypty
- 0
- Komentarze HTML
- 0
- Adresy URL w SKILL.md
- 0
Brak sygnałów. Repozytorium zmienione 19.05.2026, bez odwołań do wycofanych modeli i API.
MIT, otwarta: można używać i zmieniać, z zachowaniem informacji o autorze.
Co robi i kiedy
Tworzy, testuje i aktualizuje fiszki do aktywnego przypominania oparte na pojęciach z sesji, aby wzmocnić zapamiętywanie.
Oryginalny opis (EN)
Use when creating, testing, or updating active-recall flashcards grounded in session concepts to reinforce retention.
Jak działa
SKILL.md bez nagłówków drugiego poziomu · 55 linii, 456 słów
Podgląd SKILL.md
--- name: flashcards description: Use when creating, testing, or updating active-recall flashcards grounded in session concepts to reinforce retention. version: 1.1.0 authors: - edu-agent-skills contributors tags: [productivity, flashcards, active-recall, retention] status: stable --- # Purpose Convert session concepts into structured active-recall flashcards. Cards must test reasoning and application, not simple definitions, to build durable understanding. # Activation - Concept teaching session just completed. Learner requests cards. `spaced-repetition` needs card generation/update. Revision period (exam, interview, milestone). - **Skip if**: concept hasn't been taught yet. Goal is deep exploration → `deep-dive`. Active debugging/project work → would interrupt flow. - **Routing**: generate cards *after* understanding is confirmed. Feed into `spaced-repetition` for scheduling. Cards failed 3 times → trigger `misconception-detector`. # Inputs - Concepts/skills covered, learner's confirmed level, existing card set (if updating), error patterns from assessment skills. # Card Types - **Concept**: "What is X?" → vocabulary accuracy. - **Mechanism**: "Trace what happens when X executes." → process understanding. - **Tradeoff**: "When would you NOT use X?" → decision reasoning. - **Application**: "Given [context], which [tool/pattern] and why?" → transfer. - **Debug**: "What's wrong with this code?" → diagnostic thinking. Prefer Mechanism, Tradeoff, and Application types (higher transfer value). At least 60% of cards should be these types. # Workflow 1. **Extract** — Identify 3–7 key concepts worth card-ifying. Prioritize mechanisms, tradeoffs, application patterns. Skip long-mastered concepts. 2. **Generate** — 1–2 cards per concept. Front = question (not keyword). Back = complete model response (~100 words max). Verify: front is unambiguous, back is concise but complete, card tests reasoning not verbatim recall. 3. **Test** — Present front only; learner responds before seeing back. Self-score: Easy (fluent) / Hard (needed effort) / Failed (wrong/blank). Update interval. 4. **Handle Failures** — Failed card: re-test after 10 minutes in same session. Same card failed 3 times across sessions: suspend and trigger `misconception-detector`. 5. **Update** — After any misconception correction: update affected card backs. Never leave outdated cards in the deck. # Rules - DO: card fronts must be questions, not keywords. - DO: backs must require explanation, not one-word answers. - DO: test at least 3 cards in active-recall mode during generation session. - DO: cross-reference cards for connected concepts. - DON'T: generate all "What is X?" cards — that's definition-only bias. - DON'T: let decks grow unbounded — flag cards with ease_factor > 2.5 for 5+ sessions as mastered. - DON'T: generate cards before understanding is confirmed. - DON'T: skip active testing — generation is not the end state. # Output Card generation: topic, date, cards (front + back + type + difficulty). Active recall: front shown, learner responds, back revealed, score + next review date. Format naturally. # Checklist - [ ] 60%+ cards are Mechanism/Tradeoff/Application type. - [ ] Each card front is a clear question; back is a complete response. - [ ] At least 3 cards tested in active-recall this session. - [ ] Cards updated after any misconception correction.
Pliki w paczce (3)
- README.md 1 KB
- SKILL.md 3 KB
- examples.md 2 KB
Wersje
| Repozytorium | Wynik | Licencja |
|---|---|---|
| Ta strona · yugash007/edu-agent-skills | 90 | MIT |
| yarikleto/claude-teacher-plugin | 100 | MIT |
| ic3b3rg/learn-skills | 100 | MIT |
| anthropics/claude-for-legal | 100 | Apache-2.0 |
| uk-agents/uk-legal-plugins | 100 | Apache-2.0 |
| zekaisuni/claude-for-legal-turkish | 100 | Apache-2.0 |
| zhou210712/claude-for-legal-zh | 100 | Apache-2.0 |
| LandonSchropp/agent-toolkit | 85 | MIT |
Migawka bazy z 13.11.2026 · wersja paczki 19bac1bdbc96 (skrót SHA-256 całej paczki)
Pobrania w ostatnich 30 dniach: 0 · ranking
Zgłoś problem z tym skillem: szkodliwa treść, błędny opis albo prośba autora o usunięcie.
Podobne skille
- 100active-recall
Przygotowuje ćwiczenia z przypominania bez notatek: pytania, fiszki, klucz odpowiedzi i kolejkę powtórek z podanych materiałów. - 95retention-engine
Zadaje aktywne pytania z notatek, przygotowuje zestawy egzaminacyjne i fiszki powtórek rozłożone w czasie. - 100active-recall
Zamienia notatki i slajdy w pytania na różnych poziomach Blooma, fiszki i quiz z planem powtórek systemem Leitnera. - 90ccdvf-flashcards
Przepytuje z fiszek i pytań przed egzaminem Claude z API, Agent SDK, MCP, promptów i Claude Code. - 90spaced-repetition
Zamienia materiał do nauki w pytania i fiszki, odpytuje po jednym pytaniu i układa harmonogram powtórek dopasowany do zapominania. - 100anki
Tworzy fiszki Anki, dodaje notatki, zarządza taliami i pomaga w nauce metodą powtórek w odstępach.