The Agent Skills Directory

[SAFE]: The skill is a security auditing tool meant to analyze the safety of other skills. No malicious patterns were identified in its own implementation.
[COMMAND_EXECUTION]: The provided scripts/gate.py script performs static text analysis of local files. It uses regular expressions to identify potentially dangerous patterns in target code but does not execute the target code itself.
[DATA_EXFILTRATION]: No network operations or data transmission patterns were found. The skill works entirely offline to produce local JSON and Markdown reports.
[EXTERNAL_DOWNLOADS]: The skill does not download external scripts or dependencies. The main script relies solely on Python's standard library.
[CREDENTIALS_UNSAFE]: While the script contains regex patterns to detect hardcoded secrets (e.g., AWS_SECRET_ACCESS_KEY), these are used strictly for detection purposes when scanning target files and do not represent hardcoded credentials within the skill itself.

skill-intake-sandbox-gate