Where are the review results stored?

All findings, including status and specific issues, are written to a standardized .task/review-sonnet.json file.

When will a review be marked as 'needs_changes'?

The status is set to 'needs_changes' if any 'error' severity issue is found or if two or more 'warning' severity issues are identified.

How does it decide between a Plan or Code review?

The skill automatically detects your current progress by checking for the existence of .task/plan-refined.json or .task/impl-result.json files.

What is Review Sonnet?

Review Sonnet is a Claude Code skill that provides rapid, practical code and plan reviews focused on security, quality, and testing efficiency.

Does it check for security vulnerabilities?

Yes, it specifically audits for OWASP Top 10 issues including SQL injection, hardcoded secrets, XSS vulnerabilities, and missing authentication checks.

Review Sonnet

Name: Review Sonnet
Author: Z-M-Huang

byZ-M-Huang

•

Seguridad y Pruebas

Performs rapid, automated code and plan reviews to identify quality issues, security vulnerabilities, and testing gaps.

Acerca de

Review Sonnet is a specialized Claude Code skill designed for high-speed, practical feedback during the development lifecycle. It intelligently toggles between Plan Reviews and Code Reviews by detecting existing task artifacts, focusing on OWASP Top 10 security risks, logical correctness, and test coverage adequacy. By providing standardized JSON outputs and clear decision rules, it serves as an efficient gatekeeper that catches obvious errors and architectural gaps before deeper manual analysis is required.

Características Principales

Dual-mode review capabilities for both implementation plans and finalized code
11 GitHub stars
Speed-optimized analysis using the high-performance Claude Sonnet model
Standardized JSON reporting for seamless integration into development workflows
Automated security auditing focusing on OWASP Top 10 vulnerabilities
Comprehensive test coverage evaluation and validation

Casos de Uso

Pre-screening pull requests for obvious logic bugs and hardcoded secrets
Verifying test coverage for new features before final merging
Validating implementation plans for missing requirements or security flaws

Acerca de

Características Principales

Dual-mode review capabilities for both implementation plans and finalized code
11 GitHub stars
Speed-optimized analysis using the high-performance Claude Sonnet model
Standardized JSON reporting for seamless integration into development workflows
Automated security auditing focusing on OWASP Top 10 vulnerabilities
Comprehensive test coverage evaluation and validation

Casos de Uso

Pre-screening pull requests for obvious logic bugs and hardcoded secrets
Verifying test coverage for new features before final merging
Validating implementation plans for missing requirements or security flaws