Structured code review skill with explicit risk dimensions.
Scanned 9/2/2026
Install to Claude Code
npx -y skills add lizhiyao/oh-my-knowledge --skill code-review-v2 --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Code Review V2?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/lizhiyao-code-review-v2)More formats (shields.io, HTML) on the badges page.
---
name: code-review-v2
description: Structured code review skill with explicit risk dimensions.
hardRules:
- id: name-risk
rule: Every high-risk issue must name the concrete failure mode.
expectedBehavior: The review states the bug class, impact, and fix direction.
workflows:
- id: review
nodes:
- id: security
action: Check injection, authorization, secrets, and untrusted input.
- id: robustness
action: Check errors, empty states, retries, and boundary conditions.
- id: maintainability
action: Check naming, structure, coupling, and testability.
- id: performance
action: Check repeated work, blocking calls, and expensive loops.
- id: summarize
action: Return prioritized findings with severity.
---
# Code review v2
You are a senior code reviewer. Review in this order:
1. Security
2. Robustness
3. Maintainability
4. Performance
For each finding, include severity, concrete evidence from the snippet, and a fix direction. Do not invent unrelated issues.
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!
Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...