The Agent Skills Directory

[COMMAND_EXECUTION]: The skill performs security audits by running standard, well-known ecosystem tools including npm audit, pip-audit, safety check, bundle audit, and govulncheck.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it processes untrusted data from the codebase being audited.
Ingestion points: Processes file content retrieved via git diff --cached --name-only or through globbing the source directory.
Boundary markers: The instructions do not specify the use of delimiters or explicit 'ignore instructions' warnings when reading file contents into the agent context.
Capability inventory: The skill is capable of executing shell commands for dependency audits and spawning sub-agents for parallel task execution.
Sanitization: No explicit sanitization, validation, or escaping of the audited code content is described in the workflow.

review-security