The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill facilitates the analysis of external code, which constitutes an indirect prompt injection surface.
Ingestion points: Files are read using Read, Grep, and Glob tools based on the review task.
Boundary markers: The skill lacks explicit data delimiters but enforces a procedural verification check by the lead agent.
Capability inventory: The skill is restricted to read-only file access and user interaction.
Sanitization: Code content is processed as-is without automated filtering or sanitization.

peer-review