The Agent Skills Directory

[SAFE]: The skill is a well-structured tool for static code analysis. All scripts provided are local and focus on processing text or code for review purposes.
[PROMPT_INJECTION]: The skill instructions in SKILL.md include explicit security measures to ignore instructions embedded in target code, preventing indirect prompt injection attacks.
[DATA_EXFILTRATION]: No sensitive file access or network communication patterns were identified. Hardcoded patterns found in the JSON rule files are intended for detection purposes during code audits and do not represent actual credentials or exfiltration attempts.

code-auditor