원클릭으로
sh-debug
Structured root-cause diagnosis before proposing fixes
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
메뉴
Structured root-cause diagnosis before proposing fixes
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Multi-expert AI/ML specification review with scoring gate — model architecture, evaluation rigor, safety/alignment, and production readiness for AI systems and LLM apps
Multi-expert architecture review — boundaries, integration patterns, failure modes, evolvability
Multi-expert business analysis with advisory recommendations (no scoring gate)
Multi-expert intelligence pipeline review — discovery quality, ingestion resilience, scoring validity, platform compliance, taxonomy coherence, cost efficiency
Multi-expert mobile/native app specification review with scoring gate — Android, iOS, Swift, SwiftUI
Multi-expert personal-development review with scoring gate — learnability, adoption, human-centeredness, and capability impact for talent/learning/AI-augmentation designs
| name | sh-debug |
| description | Structured root-cause diagnosis before proposing fixes |
Diagnose bugs, test failures, and unexpected behavior using structured root-cause analysis. No fixes without understanding. Symptom fixes are failure.
NO FIXES WITHOUT ROOT CAUSE INVESTIGATION FIRST
Any technical issue: test failures, runtime bugs, unexpected behavior, performance problems. Use ESPECIALLY when under time pressure or when "just one quick fix" seems obvious.
Complete each phase before proceeding to the next.
| Phase | Key Activity | Gate |
|---|---|---|
| 1. Root Cause | Read errors, reproduce, trace | Understand WHAT and WHY |
| 2. Pattern | Find working examples, compare | Identify differences |
| 3. Hypothesis | Form theory, test minimally | Confirmed or new hypothesis |
| 4. Implement | Create test, fix, verify | Bug resolved, tests pass |
Will: Systematic diagnosis, evidence gathering, root-cause identification, targeted fix. Will not: Apply fixes without investigation, stack multiple fixes, skip reproduction.
After diagnosis, apply the fix. Use /sh:test to verify. Use /sh:analyze if issue reveals broader quality concerns.