security-scan

The security-scan skill performs automated security vulnerability analysis on codebases during review workflows. It is designed to help identify critical security risks before code is approved or deployed, ensuring safer software delivery. By executing pre-configured scanning scripts and analyzing the resulting reports, this skill provides a structured overview of vulnerabilities such as SQL injection, cross-site scripting (XSS), hardcoded secrets, and insecure dependencies. It eliminates the need for manual security inspection by integrating directly into the code review process.

The skill operates in two modes: a basic mode for fast initial scans focusing on high and medium severity issues, and an advanced mode for comprehensive analysis across all severity levels. It automatically detects the project’s programming language and runs the appropriate security tools, then categorizes findings by severity. The generated report includes counts of critical, high, and medium issues, along with detailed findings such as file locations and recommended actions. The skill reads this structured report and returns a concise summary, highlighting the most important issues while preserving full details in a generated artifact file.

This skill is primarily intended for technical leads, reviewers, and developers responsible for maintaining code quality and security. It is especially useful during pull request reviews, pre-deployment checks, and when working with security-sensitive components like authentication systems, APIs, or database interactions. By embedding security checks directly into the review workflow, it helps teams catch vulnerabilities early and enforce consistent security standards across projects.