ck:code-review

Review code changes for defects, security flaws, and requirement mismatches.

Updated May 10, 2026
One-click install
npx skills add https://github.com/dthuy62/kollet-app --skill ck-code-review-dthuy62
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ck:code-review
Source: https://github.com/dthuy62/kollet-app/tree/main/.opencode/skills/code-review
Command: npx skills add https://github.com/dthuy62/kollet-app --skill ck-code-review-dthuy62

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill prevents flawed code from shipping by reviewing changes with technical rigor, catching security holes, hidden failure modes, and mismatches against intent before merge.

Core Features & Use Cases

  • Adversarial code review: Actively tries to break implementations instead of just praising them.
  • Multiple input modes: Handles PRs, commit hashes, pending changes, recent context, and full codebase scans.
  • Review workflow support: Guides spec compliance checks, edge-case scouting, and verification before completion.
  • Use case: A team member finishes a multi-file feature and wants a hostile, evidence-based review that finds regressions, security issues, and missing requirements before release.

Quick Start

Ask the skill to review the current pending changes, a specific PR, or a commit hash.

Frequently Asked Questions about ck:code-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I review code changes for security flaws and missed requirements?

Code review for security flaws and missed requirements analyzes pending diffs and pull requests to catch defects before merge. It applies adversarial red-team analysis to actively break implementations and find hidden failure modes.

Can I use adversarial analysis to find edge cases in a multi-file feature commit?

Adversarial analysis finds edge cases in a multi-file feature commit by actively trying to break the implementation. It scouts for hidden failure modes and requirement mismatches, providing evidence-based verdicts on release readiness.

What is the best way to check spec compliance during a PR review?

PR review checks spec compliance by guiding a workflow that verifies pending changes against original intent. It scouts for edge cases and performs verification before completion to prevent flawed code from shipping.

Does this code review approach work with full codebase scans and commit hashes?

Code review works with full codebase scans and commit hashes alongside pending diffs and pull requests. It handles multi-file features, bug fixes, and release readiness checks using evidence-based verification.

Why should I use adversarial red-team analysis instead of standard code review?

Adversarial red-team analysis actively tries to break implementations rather than just praising them. It catches security holes, hidden failure modes, and mismatches against intent before merge, preventing flawed code from shipping.

When do I need an evidence-based review for release readiness?

Evidence-based review for release readiness is needed when a team member finishes a multi-file feature. It provides hostile, technical rigor to find regressions, security issues, and missing requirements before release.