caveman-review

Pass

Audited by Gen Agent Trust Hub on Apr 8, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill consists entirely of natural language instructions for formatting AI agent output. No malicious patterns, such as credential theft or unauthorized data access, were detected.
  • [NO_CODE]: There are no executable scripts, binaries, or configuration files (like package.json or requirements.txt) associated with this skill, eliminating the risk of remote code execution or privilege escalation.
  • [PROMPT_INJECTION]: The instructions do not contain attempts to bypass agent safety filters or override core system behavior. The provided commands ('stop caveman-review' or 'normal mode') are benign functional toggles for the stylistic behavior defined in the skill.
  • [DATA_EXFILTRATION]: No network-enabled tools or sensitive file paths are referenced. The skill operates purely within the text-processing context of a code review task.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 8, 2026, 11:24 PM