vulnerability-scanner

Run security_scan.py with the project path to detect hardcoded credentials including AWS keys, JWT tokens, database connection strings, and private keys. The scanner uses regex patterns across code and config files, classifying findings by severity from critical to medium.

The scan covers four areas: dependency vulnerabilities and missing lock files (A03 supply chain), hardcoded secrets (A04 cryptographic failures), dangerous code patterns like eval and SQL concatenation (A05 injection), and misconfigurations like debug mode and wildcard CORS (A02).

Execute python scripts/security_scan.py followed by your project path. Use --scan-type to limit scanning to deps, secrets, patterns, or config, and --output summary for a human-readable report instead of JSON.

Yes, the scanner analyzes files with extensions including .js, .ts, .jsx, .tsx, .py, .go, .java, .rb, and .php, plus configuration files like .json, .yaml, and .env. It also runs npm audit automatically when a package.json is present.

Findings are classified as critical, high, medium, or low based on exploitability and impact. The methodology combines CVSS base scores with EPSS exploit likelihood and asset value, so actively exploited vulnerabilities with high EPSS scores are flagged for immediate action.

Automated scans detect known patterns but cannot identify business logic flaws, insecure design issues, or complex authentication bypasses. The skill recommends combining scanner output with manual threat modeling, attack surface mapping, and the provided audit checklists.