# High Risk Review

> Apply enhanced verification (CoVe-like review, security/arch/perf checks, targeted web research) for high-risk tasks or uncertainty. Use for security, infra, data loss risk, major refactors, or when facts may be outdated.

- Skill: `janjaszczak/high-risk-review` (Agent Skill)
- Install (CLI): `npx skillmds@latest add janjaszczak/high-risk-review`
- Raw SKILL.md: https://api.skillmd.com/api/skills/janjaszczak/high-risk-review/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: DevOps & Infra
- Author: janjaszczak (https://skillmd.com/u/janjaszczak)
- Updated: 2026-09-09
- Page: https://skillmd.com/skills/janjaszczak/high-risk-review

---


# high-risk-review

## Activation gate (anti-noise)
Activate if:
- Security/auth/crypto, infra, permissions, data loss, performance hot paths.
- Uncertain or time-sensitive facts.
- User asks to “zweryfikuj / upewnij się”.

## Procedure
1. Draft solution.
2. Identify 3–5 highest-impact claims/assumptions.
3. Verify independently:
   - Repo evidence (if applicable)
   - Minimal test commands
   - Web research (authoritative sources)
   - For code security patterns: apply the checklist in `agents/security-auditor.md` (or invoke the security-auditor subagent).
4. Revise solution and mark remaining uncertainties.

## Output
- Final recommendation + explicit verification steps.