lidge-jun/codexclaw / plugins/codexclaw/skills/pabcd
cxc-pabcd
VerifiedUse for Plan-Audit-Build-Check-Done work. Triggers: PABCD, plan this, 기획, 단계별로, 요구사항 정리.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 5.2 KB |
| agents/openai.yaml | config | 147 B |
| references/catalog-discovery.yaml | config | 15.5 KB |
| references/delegation.md | markdown | 19.4 KB |
| references/dispatch-surfaces.md | markdown | 13.9 KB |
| references/implementation-units.md | markdown | 3.7 KB |
| references/loop-engineering.md | markdown | 9.3 KB |
| references/optimization.md | markdown | 3.9 KB |
| references/phase-audit.md | markdown | 8.4 KB |
| references/phase-check.md | markdown | 4.7 KB |
| references/phase-control.md | markdown | 9.7 KB |
| references/phase-plan.md | markdown | 6.5 KB |
| references/plan-output.md | markdown | 3.9 KB |
references/catalog-discovery.yaml
View on GitHub# Catalog Discovery — option ontology for the PABCD Interview (design/UX-first).
#
# WHAT THIS IS: a versioned menu of app-building choices that the `catalog_discovery`
# interview sub-mode presents to a novice who does not yet know the option space.
#
# LOAD-BEARING INVARIANT: design/UX LEADS. `axis_order` is ascending by stage;
# Stage 1 (design) is asked FIRST, Stage 2 (domain) second, and Stage 3 (feature/
# data/security/ops/cost) entries DERIVE from the Stage 1+2 answers (`derived_from`).
#
# SCOPE NOTE (audit fix #2): this YAML encodes derivation INPUTS + dependency metadata.
# It does NOT execute derivation. The agent procedure (which questions to ask, in what
# order, how to resolve `implies`/`derived_from`) lives in dev-pabcd SKILL.md prose.
# Automated runtime filtering is out of scope and would escalate to code.
version: 1
last-verified: 2026-07-04
# Ascending stage order. A conforming consumer MUST iterate this ascending and MUST NOT
# present a stage until all earlier stages are answered (enforced by SKILL.md prose).
axis_order:
- {stage: 1, axis: design, label_ko: "디자인/UX", note: "asked FIRST — most important"}
- {stage: 2, axis: domain, label_ko: "앱 유형", note: "what kind of app"}
- {stage: 3, axis: feature, label_ko: "기능", derived: true}
- {stage: 3, axis: data, label_ko: "데이터", derived: true}
- {stage: 3, axis: security, label_ko: "보안/개인정보", derived: true}
- {stage: 3, axis: ops, label_ko: "운영/자동화", derived: true}
- {stage: 3, axis: cost, label_ko: "비용/복잡도", derived: true}
# Design methodology wiring (from dev-uiux-design). Step 1 = Product-Personality Selection.
design_methodology:
primary: product_personality_selection
ref: "skills/dev-uiux-design/SKILL.md"
followups:
- step: korean_request_translation # dev-uiux-design §3 — map Korean adjectives to design tokens
- step: reference_discovery # dev-uiux-design §1 Step 6 — anchor with existing product visuals
- step: design_read # dev-uiux-design §2 — full Design Read for production-ready dials
# ============================================================================
# STAGE 1 — DESIGN (6 dials, leading). Each is `required` and carries structured
# `question_options` so the interview never improvises design questions (audit fix #4).
# ============================================================================
entries:
- id: design.mood
axis: design
stage: 1
required: true
label: {ko: "분위기", en: "Mood"}
teach_text:
ko: "앱을 열었을 때의 첫인상과 감성적 온도. 아늑한 카페 vs 유리 사무실 vs 럭셔리 쇼룸처럼 색과 서체가 경험을 예고합니다."
en: "The first impression and emotional temperature when the app opens — like a cozy cafe vs a glass office vs a luxury showroom."
examples: {ko: "Toss(친근), Stripe(세련)", en: "Toss (friendly), Stripe (refined)"}
question_options:
- {value: friendly, label_ko: "친근하고 따뜻한", tradeoff_ko: "접근성↑ 신뢰감은 별도 장치 필요", implies_token: "warm palette / rounded"}
- {value: refined, label_ko: "세련되고 프리미엄", tradeoff_ko: "고급감↑ 캐주얼한 친밀감은↓", implies_token: "muted palette / sharp"}
- {value: mystical, label_ko: "신비롭고 몰입적", tradeoff_ko: "테마 몰입↑ 정보 명료성 주의", implies_token: "dark cosmic palette"}
token_hint: "personality/palette branch"
- id: design.lightness
axis: design
stage: 1
required: true
label: {ko: "밝기", en: "Lightness"}
teach_text:
ko: "화면의 기본 톤. 밝은 모드는 대낮의 종이책처럼 명료하고, 어두운 모드는 밤의 네온사인처럼 몰입적·집중적입니다."
en: "The base tone — light is a paper book in daylight; dark is a neon sign at night, immersive and focused."
examples: {ko: "Notion(라이트), Linear(다크)", en: "Notion (light), Linear (dark)"}
question_options:
- {value: light, label_ko: "밝은 모드", tradeoff_ko: "가독성·정돈감↑", implies_token: "bg-white / text-gray-900"}
- {value: dark, label_ko: "어두운 모드", tradeoff_ko: "몰입·피로도↓, 대비 설계 필요", implies_token: "bg-gray-950 / text-gray-100"}
token_hint: "theme lightness"
- id: design.density
axis: design
stage: 1
required: true
label: {ko: "정보 밀도", en: "Density"}
teach_text:
ko: "한 화면에 담는 정보량과 여백. 여유로운 밀도는 미술관처럼 한 콘텐츠에 집중시키고, 높은 밀도는 조종석처럼 많은 정보를 한눈에 비교하게 합니다."
en: "How much fits on one screen — spacious like a museum vs dense like a cockpit."
examples: {ko: "Apple(여유), GitHub(빽빽)", en: "Apple (spacious), GitHub (dense)"}
question_options:
- {value: spacious, label_ko: "여유롭게", tradeoff_ko: "초보 친화↑ 스크롤↑", implies_token: "py-8 / gap-8 (D1-D3)"}
- {value: dense, label_ko: "정보 밀집", tradeoff_ko: "숙련자 효율↑ 첫인상 복잡", implies_token: "py-1 / gap-1 (D4-D8)"}
token_hint: "visual density scale"
- id: design.shape
axis: design
stage: 1
required: true
label: {ko: "모서리 곡률", en: "Shape"}
teach_text:
ko: "버튼·카드의 둥글기. 각진 형태는 설계도처럼 정밀·구조적이고, 둥근 형태는 젤리처럼 친근·캐주얼합니다."
en: "Corner roundness — sharp reads structural/precise, soft reads friendly/casual."
examples: {ko: "Toss(둥긂), Vercel(각짐)", en: "Toss (rounded), Vercel (sharp)"}
question_options:
- {value: rounded, label_ko: "둥근 형태", tradeoff_ko: "친근함↑", implies_token: "rounded-2xl (16px+)"}
- {value: sharp, label_ko: "각진 형태", tradeoff_ko: "정밀·기술감↑", implies_token: "rounded-none/md (0-6px)"}
token_hint: "border radius"
- id: design.typography
axis: design
stage: 1
required: true
label: {ko: "글꼴 스타일", en: "Typography"}
teach_text:
ko: "서체 종류와 대비. 명조/서예체는 역사·문학적 깊이를 주지만 모바일 가독성 주의가 필요하고, 고딕체는 가독성이 안정적입니다."
en: "Typeface character — serif adds literary depth (watch mobile legibility); sans is reliably legible."
examples: {ko: "Stripe(우아), Notion(가독)", en: "Stripe (elegant), Notion (legible)"}
question_options:
- {value: serif, label_ko: "명조/서예체", tradeoff_ko: "감성·깊이↑ 좁은 화면 가독성 주의", implies_token: "Serif display"}
- {value: sans, label_ko: "고딕체", tradeoff_ko: "가독성 안정↑ 전통미↓", implies_token: "Pretendard/sans"}
token_hint: "font family"
- id: design.motion
axis: design
stage: 1
required: true
label: {ko: "인터랙션 감도", en: "Motion"}
teach_text:
ko: "호버·전환 시 움직임의 반응. 부드러운 스프링은 감성·고급감을 주고, 즉시 반응은 빠르고 명료한 조작감을 줍니다."
en: "Motion feel — soft spring reads premium; instant reads fast and clear."
examples: {ko: "Apple(스프링), Linear(즉시)", en: "Apple (spring), Linear (instant)"}
question_options:
- {value: spring, label_ko: "부드러운 스프링", tradeoff_ko: "고급감↑ 체감 속도↓", implies_token: "spring easing"}
- {value: instant, label_ko: "즉각 반응", tradeoff_ko: "속도·명료성↑ 감성↓", implies_token: "instant/linear"}
token_hint: "motion tempo"
# ==========================================================================
# STAGE 2 — DOMAIN (app-type ontology). Chosen after design; seeds derived defaults.
# ==========================================================================
- id: domain.content_service
axis: domain
stage: 2
label: {ko: "콘텐츠 서비스", en: "Content service"}
teach_text:
ko: "사용자에게 해석·콘텐츠·결과를 제공하는 앱 (예: 사주·운세, 블로그, 미디어)."
en: "Serves interpretations/content/results (e.g. saju/fortune, blog, media)."
examples: {ko: "사주 앱, 운세, 뉴스", en: "Saju app, horoscope, news"}
implies: []
- id: domain.marketplace
axis: domain
stage: 2
label: {ko: "마켓플레이스", en: "Marketplace"}
teach_text: {ko: "판매자와 구매자를 연결하는 거래 앱.", en: "Connects sellers and buyers."}
examples: {ko: "당근, 쿠팡", en: "Danggeun, Coupang"}
implies: [feature.payments]
- id: domain.dashboard
axis: domain
stage: 2
label: {ko: "대시보드", en: "Dashboard"}
teach_text: {ko: "데이터를 모아 지표로 보여주는 관리·분석 앱.", en: "Aggregates data into metrics."}
examples: {ko: "Datadog, Stripe Dashboard", en: "Datadog, Stripe"}
implies: [security.auth]
- id: domain.booking
axis: domain
stage: 2
label: {ko: "예약", en: "Booking"}
teach_text: {ko: "시간·자원 예약을 관리하는 앱.", en: "Manages time/resource reservations."}
examples: {ko: "네이버 예약, Calendly", en: "Naver Booking, Calendly"}
implies: [ops.scheduled_jobs]
- id: domain.community
axis: domain
stage: 2
label: {ko: "커뮤니티", en: "Community"}
teach_text: {ko: "사용자 게시물·상호작용 중심 앱.", en: "User posts and interaction."}
examples: {ko: "Reddit, 디시", en: "Reddit, DCinside"}
implies: [data.user_generated, ops.admin_review]
- id: domain.ai_agent
axis: domain
stage: 2
label: {ko: "AI 에이전트", en: "AI agent"}
teach_text: {ko: "AI가 핵심 가치를 만드는 앱.", en: "AI is the core value."}
examples: {ko: "ChatGPT, Perplexity", en: "ChatGPT, Perplexity"}
implies: [cost.infra_complexity]
- id: domain.internal_tool
axis: domain
stage: 2
label: {ko: "내부 도구", en: "Internal tool"}
teach_text: {ko: "팀 내부 업무를 처리하는 도구.", en: "Handles internal team workflows."}
examples: {ko: "Retool, Notion", en: "Retool, Notion"}
implies: [security.auth]
# ==========================================================================
# STAGE 3 — DERIVED. Each carries `derived_from` (Stage 1/2 ids) so the interview
# surfaces it FROM earlier answers rather than as a flat list.
# ==========================================================================
- id: security.pii_protection
axis: security
stage: 3
derived_from: [domain.content_service, domain.community]
label: {ko: "개인정보 보호", en: "Personal data protection"}
teach_text:
ko: "생년월일·위치 등 민감정보를 수집하면 암호화·접근통제·보존정책이 필요합니다. 사주 앱의 생시(生時)가 대표적입니다."
en: "Collecting birthdays/locations requires encryption, access control, retention policy."
examples: {ko: "건강 앱, 사주 앱", en: "Health apps, fortune apps"}
implies: [data.retention_policy, ops.scheduled_jobs]
risk_class: high
cost_class: 2
auto_activate_rules: ["생년월일", "생시", "birth", "사주", "personal", "위치"]
- id: security.auth
axis: security
stage: 3
derived_from: [domain.dashboard, domain.internal_tool, feature.payments]
label: {ko: "인증/로그인", en: "Authentication"}
teach_text: {ko: "사용자를 식별하고 세션을 안전하게 유지합니다.", en: "Identify users and keep sessions safe."}
examples: {ko: "OAuth, 이메일 로그인", en: "OAuth, email login"}
risk_class: high
cost_class: 2
- id: data.retention_policy
axis: data
stage: 3
derived_from: [security.pii_protection]
label: {ko: "보존/파기 정책", en: "Retention policy"}
teach_text: {ko: "회원 탈퇴·미접속 시 개인정보를 언제 파기할지 정합니다.", en: "When to delete personal data on churn/inactivity."}
examples: {ko: "탈퇴 즉시 파기, 1년 미접속 파기", en: "Delete on churn, 1yr-inactive purge"}
implies: [ops.scheduled_jobs]
risk_class: high
cost_class: 1
auto_activate_rules: ["파기", "탈퇴", "retention", "보존"]
- id: data.user_generated
axis: data
stage: 3
derived_from: [domain.community, domain.marketplace]
label: {ko: "사용자 생성 콘텐츠", en: "User-generated content"}
teach_text: {ko: "사용자가 올리는 글·이미지의 저장·검증·신고 처리가 필요합니다.", en: "Store/validate/moderate user uploads."}
examples: {ko: "게시판, 리뷰", en: "Boards, reviews"}
implies: [ops.admin_review]
risk_class: medium
cost_class: 2
- id: feature.payments
axis: feature
stage: 3
derived_from: [domain.marketplace, domain.content_service]
label: {ko: "결제", en: "Payments"}
teach_text: {ko: "유료 기능·구독을 받으면 PCI·환불·정산 흐름이 필요합니다.", en: "Charging requires PCI/refund/settlement flows."}
examples: {ko: "구독, 인앱결제", en: "Subscription, IAP"}
implies: [security.auth]
risk_class: high
cost_class: 3
auto_activate_rules: ["결제", "구독", "payment", "유료"]
- id: feature.notifications
axis: feature
stage: 3
derived_from: [domain.content_service, domain.booking]
label: {ko: "알림", en: "Notifications"}
teach_text: {ko: "푸시·이메일 알림은 발송 스케줄과 대기열이 필요합니다 (예: 매일 '오늘의 사주').", en: "Push/email needs scheduling + queues."}
examples: {ko: "오늘의 운세 푸시", en: "Daily fortune push"}
implies: [ops.scheduled_jobs]
risk_class: low
cost_class: 2
auto_activate_rules: ["알림", "푸시", "notification", "매일"]
- id: ops.scheduled_jobs
axis: ops
stage: 3
derived_from: [feature.notifications, data.retention_policy, security.pii_protection, domain.booking]
label: {ko: "예약 작업", en: "Scheduled jobs"}
teach_text: {ko: "지정 시간에 자동 실행되는 백그라운드 작업 (예: 매일 오전 8시 알림, 자정 데이터 정리).", en: "Time-triggered background routines."}
examples: {ko: "cron, 스케줄러", en: "cron, scheduler"}
implies: [ops.task_queues]
risk_class: low
cost_class: 2
auto_activate_rules: ["cron", "schedule", "매일", "정기"]
- id: ops.task_queues
axis: ops
stage: 3
derived_from: [ops.scheduled_jobs]
label: {ko: "작업 큐", en: "Task queues"}
teach_text: {ko: "무거운 작업을 비동기로 처리하는 대기열.", en: "Async queue for heavy work."}
examples: {ko: "BullMQ, SQS", en: "BullMQ, SQS"}
risk_class: low
cost_class: 2
- id: ops.admin_review
axis: ops
stage: 3
derived_from: [data.user_generated, domain.community]
label: {ko: "관리자 검수", en: "Admin review"}
teach_text: {ko: "게시물 승인·신고 처리를 위한 관리자 화면과 흐름.", en: "Admin surface for approval/moderation."}
examples: {ko: "신고 처리, 승인 대기", en: "Report handling, approval queue"}
risk_class: medium
cost_class: 2
- id: cost.infra_complexity
axis: cost
stage: 3
derived_from: [domain.ai_agent, feature.payments, ops.scheduled_jobs]
label: {ko: "인프라 복잡도", en: "Infra complexity"}
teach_text: {ko: "선택한 기능들이 만드는 운영·비용 부담의 총합 지표.", en: "Roll-up of ops/cost burden from chosen features."}
examples: {ko: "GPU 비용, 다중 서비스", en: "GPU cost, multi-service"}
risk_class: medium
cost_class: 3
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |