davidcoachdev/dcfiles/check
Claude Code and agent skill by davidcoachdev/dcfiles
- community
Checking that the SKILL.md file is still in the repository.
Fase Check — gap analysis + peer review + security axis. Trigger: "/sdd-cavekit check", "verify", "review", "gap analysis"
Install
npx vskill@latest install davidcoachdev/dcfiles/checkSKILL.md
The same SKILL.md was published earlier as davidcoachdev/dcfiles/check.
Check Phase — Gap Analysis + Peer Review (Gold)
Recibes: implementation + kits. Produces: verdict + gaps + Result Contract.
Principios gold
- Security es gate de bloqueo: el eje security (
security-auditor) corre SIEMPRE. P0/P1 → REJECT del loop. - Human-in-the-Loop: el verdict final lo decide el humano. El orchestrator presenta gaps; el user aprueba.
- Anti-patterns: revisá patrones/anti-patterns (referencia
refactoring,design-patterns). - Strict TDD check: confirmá que los tests cubren los AC y la cobertura no bajó.
Gap Analysis
Para cada requirement: Complete / Partial / Missing / Over-built.
Peer Review
Segundo agente. "Find what the builder missed, NOT agree." 6 modos: Diff Critique, Design Challenge, Threaded Debate, Delegated Scrutiny, Deciding Vote, Coverage Audit.
Multi-Axis Review (Security + Web Perf)
- Security axis →
security-auditor(OWASP Top 10 + LLM Top 10: prompt injection, excessive agency, límites de token/rate/recursion). Cubre los Security Gates de la plantilla gold. - Web Perf axis →
web-performance+core-web-vitals(Metric-Honesty Rule, severidad por CWV).
P0/P1 → REJECT.
Severity Levels
| Severity | Behavior | |
|---|---|---|
| P0 (critical) | Block advancement | |
| P1 (high) | Block advancement | |
| P2 (medium) | Logged, no block | |
| P3 (low) | Logged, no block |
Result Contract (Output)
El orchestrator lee Verdict + Critical Gaps y feedback al próximo make.
Trace
Al emitir el verdict, APPENDÉ a context/impl/trace.md una línea: Verdict: {APPROVE|REVISE|REJECT} — gaps: {…}. Así la traza por paso queda completa (build + verdict) para auditoría y steering.
Eval opcional (Patrón #5)
Si hay candidatos alternativos, podés rankearlos con cavekit-eval (pairwise + Bradley-Terry). No es obligatorio; úsalo para desempatar críticas.
Auto-Load
- peer-review
- impl-tracking
- revision (fix gaps upstream)
- security-auditor
- web-performance
- core-web-vitals
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingconfig/.config/opencode/skills/sdd-cavekit/check/SKILL.md
Popularity
- 7-day trend score
- 5
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/davidcoachdev/dcfiles/check)