ck:code-review

Review code changes across PRs, commits, and codebase scans.

1|Updated Mar 5, 2026
One-click install
npx skills add https://github.com/hotriluan/alkana_web --skill ck-code-review-hotriluan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ck:code-review
Source: https://github.com/hotriluan/alkana_web/tree/main/.opencode/skills/code-review
Command: npx skills add https://github.com/hotriluan/alkana_web --skill ck-code-review-hotriluan

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Review code quality with adversarial rigor. It ensures findings are evidence-based, and verification gates prevent premature completion claims.

Core Features & Use Cases

  • Adversarial, evidence-based reviews of changes to identify security holes, false assumptions, and failure modes.
  • Supports multiple input modes: pending changes, PR numbers, commit hashes, and full codebase scans.
  • Structured three-stage workflow (scout, code-review, adversarial) with optional task-based pipelines for complex changes.

Quick Start

Provide the target diff or PR context to start the review cycle.

Frequently Asked Questions about ck:code-review

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

FAQPage Schema
How do I perform an adversarial code review on a pull request?

Perform an adversarial code review by supplying a PR number or commit hash to trigger a three-stage workflow: scout, code-review, and adversarial. This identifies security holes and false assumptions through evidence-based checks with verification gates.

What is an adversarial code review and when do I need it?

An adversarial code review rigorously identifies security holes, false assumptions, and failure modes using evidence-based findings. You need it when standard reviews miss edge cases and require verification gates to prevent premature completion claims on complex changes.

Can I review pending changes or scan a full codebase for quality issues?

Yes, you can review pending changes or scan a full codebase. The review operates in multiple input modes including PR numbers, commit hashes, --pending changes, and full codebase scans to enable targeted or broad code quality checks.

How do I verify code quality findings and prevent false assumptions?

Verify code quality findings using a structured three-stage workflow with verification gates. This ensures all identified issues are evidence-based and prevents premature completion claims during the scout, code-review, and adversarial phases.

What is the best way to automate red-team diff analysis on commits?

Automate red-team diff analysis by feeding commit hashes or PR numbers into the task-based pipelines. This executes a structured three-stage workflow to deliver brutally honest, evidence-based reviews of failure modes and security vulnerabilities.

Does this code review approach work with task-based pipelines for complex changes?

Yes, this code review approach works with task-based pipelines for complex changes. It enforces a three-stage workflow with optional task-based pipelines and verification gates to ensure findings are evidence-based and adversarially rigorous.