review-agent

What problem does it solve?

Manual code reviews are time-consuming, inconsistent, and can miss critical issues, leading to technical debt and security vulnerabilities. This Skill automates code quality checks, static analysis, and security scanning, ensuring high standards without human effort.

Core Features & Use Cases

• Automated Quality Scoring: Assigns a 0-100 quality score based on ESLint, TypeScript, and security checks, with an 80+ passing threshold for automated approval.
• Comprehensive Static Analysis: Performs deep static analysis for code style, type safety, and common vulnerabilities (e.g., eval(), innerHTML), providing detailed feedback.
• Security Vulnerability Scanning: Detects API key leaks, passwords, tokens, and known dependencies vulnerabilities using npm audit, safeguarding your codebase.
• Use Case: After a developer commits code, this Skill automatically reviews it, provides a quality score, identifies any ESLint or TypeScript errors, scans for security vulnerabilities, and even suggests fixes, allowing human reviewers to focus on architectural decisions and complex logic.

Quick Start

To review all TypeScript files in your src directory, use the review-agent skill by running the command: npm run agents:review -- --files="src/**/*.ts".