The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/security_scan.py executes the npm audit command locally using the subprocess module to identify vulnerabilities in project dependencies. This is a standard and expected function for a security scanning tool.
[PROMPT_INJECTION]: The skill processes untrusted data by reading files from a user-provided project directory for static analysis.
Ingestion points: Code and configuration files read from the filesystem in scripts/security_scan.py.
Boundary markers: None; the script reads file contents directly into memory for regex analysis.
Capability inventory: File read access and local command execution (npm audit) via subprocess in scripts/security_scan.py.
Sanitization: The script uses regular expressions to identify patterns and does not execute or evaluate the content of the files being scanned.

vulnerability-scanner