When should I use Review Opus instead of a standard review?

Use Review Opus for high-stakes changes, complex architectural shifts, or when you require a deep security audit that identifies issues beyond surface-level linting.

How does it evaluate test quality?

It checks for coverage depth across all code paths, ensures edge cases are tested, and verifies that assertions follow FIRST principles for meaningful testing.

What specific security issues does Review Opus look for?

It analyzes code for business logic flaws, race conditions (TOCTOU), information disclosure, and authorization bypasses such as IDOR and path traversal.

Does this skill review implementation plans or final code?

It handles both. The skill automatically detects if a plan or an implementation result exists in your workspace and adjusts its review criteria to match the context.

What model does Review Opus use?

This skill is specifically configured to utilize the Claude 3.5 Opus model to leverage its superior reasoning and deep analysis capabilities for complex codebases.

Review Opus (Deep Code Review)

Name: Review Opus (Deep Code Review)
Author: cskwork

bycskwork

0•

安全与测试

Provides deep architectural analysis and security-focused code reviews using the reasoning power of Claude Opus.

Review Opus is a specialized skill designed for high-stakes code and plan analysis, leveraging the advanced reasoning capabilities of the Claude Opus model. It functions as a rigorous secondary review layer that goes beyond surface-level syntax, specifically targeting subtle logic flaws, race conditions, architectural technical debt, and test coverage depth. Whether reviewing a technical plan before implementation or auditing the final code, this skill ensures that software designs are scalable, secure, and maintainable by identifying complex vulnerabilities that quick reviews often miss.

主要功能

01Deep architectural analysis for long-term maintainability and scalability

02Automated plan reviews to validate technical approaches before coding starts

03Exhaustive test quality assessment against FIRST principles

040 GitHub stars

05Comprehensive security audits for race conditions and authorization flaws

06Structured JSON reporting for seamless integration with CI/CD workflows

使用场景

01Conducting final security audits for critical production-bound features

02Validating complex system designs to prevent architectural over-engineering

03Auditing codebases for subtle concurrency bugs and performance bottlenecks

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add cskwork/llm-review review-opus

For use in Claude.ai and ChatGPT