skills/juliusbrussee/cavekit/review/Gen Agent Trust Hub

review

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or data exfiltration attempts were detected. The skill uses professional instructional language to define its functional scope.
  • [COMMAND_EXECUTION]: The instructions suggest using the grep tool to read codebase modules. This is a standard and safe operation within a developer tool context, used here to verify that the specification aligns with existing code patterns and invariants.
  • [EXTERNAL_DOWNLOADS]: The skill mentions fetching "live" best-practice information. This involves standard network-based information retrieval (such as searching documentation or technical blogs) and does not involve the download or execution of unverified scripts or binaries.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 06:43 AM
Security Audit — agent-trust-hub — review