The Agent Skills Directory

[SAFE]: The skill is a logic-heavy classification engine that strictly adheres to its defined scope of categorizing code review findings.
[SAFE]: No obfuscation, prompt injection, or malicious persistence mechanisms were detected.
[SAFE]: Although the skill has access to tools like Bash and WebFetch, the instructions explicitly forbid code modifications and focus entirely on data processing and classification.
[SAFE]: The potential for indirect prompt injection via processed project data is addressed by the skill's constrained analytical focus and prohibitions on refactoring.

review-implementation