The Agent Skills Directory

[SAFE]: The skill is designed to manage authorization logic, which is inherently sensitive. However, its instructions prioritize safety by mandating the detection of existing project patterns and forbidding unilateral implementation of security systems without user consent.
[COMMAND_EXECUTION]: The skill employs rg (ripgrep) to search for security-related strings in the local codebase. This command is executed in a read-only manner and uses hardcoded search patterns, preventing command injection or data exfiltration.
[PROMPT_INJECTION]: No attempts to bypass agent safety filters or override system instructions were found. The skill includes strong constraints to prevent autonomous, silent modifications to critical security infrastructure.
[DATA_EXFILTRATION]: No network operations (such as curl, wget, or fetch) or hardcoded credentials were detected. Data handling is limited to local codebase discovery.

rbac-support