메뉴
Multi-lens adversarial review with Skeptic, Architect, and Minimalist lenses and 7+ attack vectors
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
Use when implementing features, fixing bugs, refactoring code, optimizing performance, resolving deprecations, or any multi-step development task requiring planning, implementation, testing, and review. Orchestrates specialized agent teammates through research, architecture, coding, QA, code review, adversarial review, and documentation phases. Triggers on: "implement", "build", "fix bug", "refactor", "add feature", "develop this", "help me build", "add functionality", "optimize performance", "resolve deprecation", "systematic development". Do NOT trigger on: simple questions ("what does this code do?"), file searches ("where is the auth function?"), one-off commands ("run the tests"), code explanations, quick edits, or non-development tasks.
Auto-improve any agent prompt using Karpathy's autoresearch method with iterative test-measure-improve loops
Safety guardrail that blocks destructive commands for the current session
Core development rules and philosophy for coding standards, git practices, and quality guidelines
Directory freeze guardrail that restricts file edits to a specific directory for the current session
Comprehensive security review skill for authentication, user input, secrets, API endpoints, and OWASP Top 10
| name | adversarial-review |
| description | Multi-lens adversarial review with Skeptic, Architect, and Minimalist lenses and 7+ attack vectors |
Perform multi-lens adversarial review to challenge implementation correctness, structural fitness, and necessity. Apply Skeptic, Architect, and Minimalist lenses with attack vectors (V1-V8) and a Destructive Action Gate. Produce a verdict (PASS/CONTESTED/REJECT).
Triggers on: "/super-dev:adversarial-review", Stage 10 invocation by Team Lead
Determine what to review. State intent. Assess change size for reviewer count (Small: 1 Skeptic, Medium: 2 Skeptic+Architect, Large: 3 all lenses). Challenge correctness — V1 false assumptions, V2 edge cases, V3 failure modes, V4 adversarial input, V5 safety/compliance, V6 grounding audit, V8 behavior coverage. Challenge structural fitness — V1 architectural assumptions, V3 graceful degradation, V5 security boundaries, V7 dependencies. Challenge necessity — V7 dependency justification, unnecessary abstractions, premature configuration. Always-on scan for: DAT (data destruction), IRR (irreversible state), PRD (production impact), PRM (permission escalation), SEC (secret operations). No safeguard → HALT finding. HALT findings → CONTESTED minimum (multiple HALTs → REJECT). No high-severity → PASS. High with disagreement → CONTESTED. High with consensus → REJECT. Verdict only — do NOT make code changes Every finding MUST include file:line references and concrete recommendations Each lens is exclusive — no blending between reviewer perspectives HALT findings cannot be downgraded; require explicit Team Lead acknowledgment