The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes memory entries which contain untrusted user data. While this presents an indirect prompt injection surface, the skill's operations are limited to string comparison and metadata analysis for reporting purposes. It does not execute, interpret, or interpolate this data into commands, mitigating the risk of exploitation.
[SAFE]: The skill follows least-privilege principles by explicitly stating it is read-only and scoping data access to the active user and project IDs.
[SAFE]: No obfuscation, hardcoded credentials, or unauthorized network operations were detected. All described behavior is consistent with the skill's stated purpose.

memory-reviewer