How does this skill differ from a standard linter?

While linters focus on syntax and style, this skill analyzes logical flow, runtime assumptions, and complex patterns like race conditions that static analysis often misses.

Can it detect platform-specific issues?

Yes, it includes checks for production-specific concerns like timezone-naive code, locale dependencies, and floating-point precision issues.

What languages does this skill support?

It is designed to work across most major programming languages, with a particular focus on logic patterns common in JavaScript, TypeScript, Python, and similar high-level languages.

Is it suitable for security audits?

Yes, by identifying TOCTOU vulnerabilities and improper error handling, it serves as a valuable tool for finding security-relevant correctness bugs.

Code Review Correctness

Name: Code Review Correctness
Author: xinbenlv

byxinbenlv

安全与测试

Analyzes source code to identify complex logic bugs, unhandled edge cases, and critical error handling failures before they reach production.

关于

The Code Review Correctness skill acts as a senior 'Logic Bug Hunter' within Claude Code, specifically designed to surface deep-seated runtime issues that automated linters often miss. It meticulously checks for off-by-one errors, race conditions in asynchronous code, null pointer dereferences, and improper resource management. By applying a rigorous checklist based on thousands of production incident patterns, it helps developers ensure their core business logic is robust, idempotent, and capable of handling unexpected inputs or environment-specific edge cases like timezone or floating-point discrepancies.

主要功能

Verification of error propagation, resource cleanup, and 'swallowed' exceptions
0 GitHub stars
Detection of complex logic errors like off-by-one and fence-post bugs
Identification of unhandled null/undefined values and type coercion issues
Edge case testing for boundary values, empty inputs, and large datasets
Deep analysis of asynchronous race conditions and shared state modifications

使用场景

Performing pre-deployment audits to ensure comprehensive error handling and resource management
Reviewing critical business logic or complex algorithms where correctness is paramount
Hardening asynchronous backend services against race conditions and state corruption

关于

主要功能

Verification of error propagation, resource cleanup, and 'swallowed' exceptions
0 GitHub stars
Detection of complex logic errors like off-by-one and fence-post bugs
Identification of unhandled null/undefined values and type coercion issues
Edge case testing for boundary values, empty inputs, and large datasets
Deep analysis of asynchronous race conditions and shared state modifications

使用场景

Performing pre-deployment audits to ensure comprehensive error handling and resource management
Reviewing critical business logic or complex algorithms where correctness is paramount
Hardening asynchronous backend services against race conditions and state corruption