The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/security_scan.py utilizes the subprocess module to execute npm audit. This is used to perform dependency vulnerability analysis as described in the skill's purpose. The command is invoked using a list of arguments, which is a standard security best practice to prevent shell injection.
[PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it ingests and processes untrusted files from a user-provided project directory. Maliciously crafted content within those files (e.g., in code comments or metadata) could attempt to influence the agent's interpretation of scan results.
Ingestion points: The scripts/security_scan.py script reads file content from the local file system using open().read() and open().readlines() in the scan_secrets and scan_code_patterns functions.
Boundary markers: Absent; the script reads file content directly for regex matching without specific delimiters to separate data from instructions.
Capability inventory: subprocess.run() is used in scripts/security_scan.py to call external audit tools.
Sanitization: Absent; the script performs regex-based pattern matching but does not sanitize or escape the content of the files before including snippets in its JSON output report.

vulnerability-scanner