본문으로 건너뛰기
cccskills
무료GitHub에 공개

revision-mode

Use when a learner needs a structured review session prioritizing high-value topics by recency, error history, and upcoming deadlines.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

설치 방법 보기

포함된 파일(3)

  • SKILL.md3.6 KB
  • examples.md1.7 KB
  • README.md764 B

SKILL.md(원문)

설치하기 전에 에이전트에게 전달되는 지시 내용을 확인할 수 있습니다.

Purpose

Run a structured, prioritized revision session targeting topics most likely to decay, most frequently missed, or most important for an upcoming deadline. Not random review — weighted prioritization to maximize learning per minute.

Activation

  • Upcoming deadline (interview, exam, demo) within 1–5 sessions. Learner requests review after a learning period. weak-area-tracker shows accumulated weak areas. 3+ sessions since last review.
  • Skip if: learner needs new material first. Goal is project implementation → build-with-me. Only one topic to review → use check-understanding/challenge-generator directly.
  • Routing: pull priorities from weak-area-tracker. Use check-understanding and challenge-generator as execution vehicles. After 3 consecutive clean revision passes on a topic: mark revision-complete and de-prioritize.

Inputs

  • Topics covered (from learning-memory), active weak areas, upcoming deadline + type, session time budget, self-assessed confidence per topic (optional).

Priority Scoring (1–10)

  • Weak area severity: 0–3 (from tracker). Recency (days since last review): 0–3. Deadline relevance: 0–2. Low self-reported confidence: 0–2.
  • Score 7+ = must-revise. Score 4–6 = should-revise. Below 4 = defer.

Workflow

  1. Scope — Gather topics + weak areas. Ask: "Deadline? Least confident topics?" Apply priority scoring.
  2. Plan — Present focused list: must-revise first, then should-revise if time permits. Cap at 4–5 topics max. State approach per topic: concept check / challenge / flashcard drill.
  3. Recap — For each topic: ask learner to summarize in 2–3 sentences. Don't re-explain unless summary reveals a gap. 2–3 min cap per recap.
  4. Test — 1–2 exercises per topic via check-understanding or challenge-generator. Mix modes. Run 3–5 due flashcards per topic if available.
  5. Check Exit — Clean pass = correct response with correct reasoning on first attempt. After 3 clean passes across sessions: exit rotation. Below 50% pass rate this session: escalate to misconception-detector.
  6. Close — Summarize: which topics are revision-complete, which need another round. Update weak-area-tracker. Recommend next session if deadline still approaching.

Rules

  • DO: revision is active — learner produces answers, not just reads.
  • DO: include weak areas in every revision session until resolved.
  • DO: state clean-pass criteria before the session, not after.
  • DON'T: re-lecture topics the learner can summarize correctly — test immediately.
  • DON'T: pack more than 5 topics per session.
  • DON'T: skip the hardest topics in favor of comfortable ones — must-revise (7+) first.
  • DON'T: let same topics appear indefinitely — enforce exit criteria.

Output

Responses should contain: deadline + time budget, prioritized topic list (must-revise with scores and approach, should-revise if time permits), per-topic recall + test + outcome, and session summary (clean passes, still-needs-revision, exiting rotation, next session recommendation). Format naturally.

Checklist

  • Priority scoring applied before session plan.
  • Must-revise topics addressed first.
  • Every topic has at least one active retrieval exercise.
  • Weak-area-tracker updated at session close.

리뷰

아직 리뷰가 없습니다. 사용해 본 소감을 남겨 주세요.

댓글

로그인 없이 쓸 수 있습니다. 사용 후기, 질문, 작동하지 않은 환경 등을 편하게 남겨 주세요.

아직 댓글이 없습니다. 첫 댓글을 남겨 보세요.

보내면 바로 공개됩니다. 이메일 주소나 전화번호 같은 개인 정보는 쓰지 마세요. URL은 2개까지입니다.

0 / 1,000자

이 양식은 reCAPTCHA로 보호되며 Google 개인정보처리방침 및 서비스 약관이 적용됩니다.

같은 저장소의 스킬

Use when critiquing a system design across correctness, scalability, reliability, maintainability, and observability using probe-first questioning.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

yugash007/edu-agent-skills102026년 5월 19일 업데이트

Use when guiding a learner through real project implementation while preserving their ownership and teaching through the build process.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

yugash007/edu-agent-skills102026년 5월 19일 업데이트

Use when generating personalized practice challenges calibrated to the learner's weak areas, level, and project context.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

yugash007/edu-agent-skills102026년 5월 19일 업데이트

Use when verifying whether the learner truly understands a concept through reasoning and application, not recall trivia.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

yugash007/edu-agent-skills102026년 5월 19일 업데이트

Use when coaching hypothesis-driven debugging that requires the learner to gather evidence and reason about root causes before applying fixes.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

yugash007/edu-agent-skills102026년 5월 19일 업데이트

deep-dive

무료

Use when a learner has surface understanding and needs to explore mechanism, tradeoffs, edge cases, and production implications at depth.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

yugash007/edu-agent-skills102026년 5월 19일 업데이트

yugash007의 스킬 모두 보기

이 스킬의 문제 신고