The Agent Skills Directory

[SAFE]: The skill is a defensive utility intended to identify and block security anti-patterns such as SQL injection, XSS, and command injection before they are committed to code.
[SAFE]: The mention of 'eval' in the reference documentation is part of a decision matrix for security warnings and does not represent a functional code injection vulnerability in the skill itself.
[SAFE]: Local state is maintained within the standard '.claude' directory for session-based caching, which prevents redundant warnings without exposing sensitive information.
[SAFE]: No external network dependencies or remote code execution patterns were identified; the skill relies on standard libraries for local processing.
[NO_CODE]: The skill references an external script 'security_reminder_hook.py' for its primary logic, but this file is not included in the provided skill package.

security-guidance