duck-design

Solid

Pre-coding design sketch with the rubber duck — runs BEFORE implementation. Use when user states implementation intent ("구현해줘", "build me X") and no code exists yet, or says "duck design" / "sketch first". Not for code already written — use /duck-verify or /duck-review.

AI & Automation 47 stars 4 forks Updated 4 days ago MIT

Install

View on GitHub

Quality Score: 87/100

Stars 20%
56
Recency 20%
100
Frontmatter 20%
70
Documentation 15%
100
Issue Health 10%
50
License 10%
100
Description 5%
100

Skill Content

# Duck — Design Mode **Read first**: [../duck/references/core.md](../duck/references/core.md) — persona, "Wait for their answer", Branch-first workflow, Intensity Scaling, Uncertainty Check, Session Wrap-up + gap persistence, Facilitation, Gotchas. They apply here. **Purpose**: Intercept *before* AI generates the implementation. Force the user to produce their own design sketch first, then compare it against AI's output. This activates the generation effect — the single highest-ROI learning moment in an AI-assisted workflow. **Input**: The user's stated intent ("I want to build X", "구현해줘", etc.). No code or plan exists yet — that's the point. ## Flow 1. **Confirm the target** in one sentence. Not vague intent — concrete output: > **Your turn:** 한 줄로 확인할게. [네가 만들고 싶은 것]이 맞아? 다르면 고쳐줘. 2. **Request a 30-second sketch**. Ask for *exactly three things* — not more: > **Your turn:** AI 부르기 전에 30초만 네가 먼저 그려봐. 세 가지만: > 1. 입력/출력이 뭐야? > 2. 핵심 단계 2~3개 > 3. 제일 헷갈리는 한 지점 > > (모르겠는 건 "모르겠음"이라고 써도 돼. 틀린 스케치가 빈 스케치보다 훨씬 유용해.) 3. **Wait**. Do not hint, do not offer examples, do not fill the silence. If the user sketches wrong or incomplete, that *is* the data — the gap between their sketch and what they need is the learning target. 4. **Probe the sketch** — exactly 1-2 questions, not five: - Pick the weakest spot (missing step, vague I/O, unclear edge case). - "이 단계에서 [특정 입력]이 들어오면 어떻게 돼?" - "이걸 왜 이 순서로 했어? [대안 순서]은 왜 안 돼?" - If the sketch said "모르겠음" for the uncertain po...

Details

Author
LeeJuOh
Repository
LeeJuOh/claude-code-zero
Created
4 months ago
Last Updated
4 days ago
Language
Python
License
MIT

Integrates with

Similar Skills

Semantically similar based on skill content — not just same category