skills.rest
salazarsebas
salazarsebasCOMMUNITY
Explore all skills by this author →
Website𝕏@salazarsebas_crGitHub

code-review

Identify code changes and assess them with layered adversarial reviews.

9|66|Updated Oct 13, 2025
Search Tags:#automation#diff#edge-cases#triage#code-review#adversarial
One-click install
npx skills add https://github.com/salazarsebas/Cougr --skill code-review-salazarsebas
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/salazarsebas/Cougr/tree/main/.claude/skills/code-review
Command: npx skills add https://github.com/salazarsebas/Cougr --skill code-review-salazarsebas

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Adversarial, structured code review reduces blind spots by running parallel review layers (Blind Hunter, Edge Case Hunter, Acceptance Auditor) and triaging findings into actionable categories.

Core Features & Use Cases

  • Parallel reviewer roles for comprehensive diffs and targeted checks.
  • Structured triage and actionable outputs suitable for integration with story/patch workflows.
  • Use Case: when reviewing a PR, surface critical issues across design, correctness, and edge cases to guide remediation.

Quick Start

Execute the code-review workflow to trigger the triage process and produce structured findings.

Frequently Asked Questions about code-review

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

FAQPage Schema
How do I automate adversarial code review for pull requests?▼

Adversarial code review automates pull request analysis by applying parallel review layers to surface critical design and correctness issues. It triages findings into actionable categories for structured remediation across diffs and commits.

What is the best way to find edge cases in a code diff?▼

Finding edge cases in a code diff requires targeted adversarial analysis. The Edge Case Hunter layer specifically evaluates changes to surface hidden vulnerabilities and boundary issues, ensuring comprehensive coverage before merging.

Can I use automated code review for spec-guided changes across different projects?▼

Automated code review supports spec-guided changes across multiple projects by applying deterministic triage rules. It uses required frontmatter contexts and diff construction rules to ensure consistent, reviewable outcomes for various workflows.

How does structured code triage work for commit reviews?▼

Structured code triage works by running parallel reviewer roles like Blind Hunter and Acceptance Auditor on commits. This layered approach categorizes findings, ensuring traceable outputs that guide developers through targeted remediation steps.

Do I need specific dependencies to run adversarial code reviews?▼

Adversarial code reviews require no specific external dependencies. The workflow operates independently, using diff construction rules and optional resources to drive consistent outcomes, making it suitable for immediate integration.

Related Skills

Explore More Software Engineering Skills→
tony-zhelonkintony-zhelonkin
1

skill-creator

Create, evaluate, and refine Claude skills with SKILL.md and eval prompts.

Community
Advanced
FernanSuozaFernanSuoza

sec-static-code-analysis

Convert raw scanner output into triaged security findings with SARIF.

Community
Advanced
YangYuS8YangYuS8

openspec-verify-change

Validate change implementations against specs, tasks, and design documents.

Community
Advanced
bmad-code-orgbmad-code-org
8

bmad-os-review-pr

Review GitHub pull requests with adversarial and edge-case analysis.

Official
Advanced