caveman-compress

Pass

Audited by Gen Agent Trust Hub on May 20, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: No malicious override or bypass attempts were detected. The instructions define a specific linguistic persona for efficiency without attempting to subvert agent safety guidelines.
  • [DATA_EXFILTRATION]: No network operations, hardcoded credentials, or sensitive file path access were identified. The skill operates entirely on text formatting.
  • [OBFUSCATION]: A scan for hidden characters (zero-width, Unicode tags), homoglyphs, and encoded strings (Base64, hex) yielded no findings.
  • [REMOTE_CODE_EXECUTION]: The skill does not perform any remote code downloads or installations. The allowed-tools field is empty, further restricting any automated tool-based risks.
  • [INDIRECT_PROMPT_INJECTION]: While the skill processes user queries and technical data, it has no tool access (allowed-tools: []) to execute commands, making it a low-risk linguistic wrapper with no exploitable surface.
Audit Metadata
Risk Level
SAFE
Analyzed
May 20, 2026, 01:33 PM
Security Audit — agent-trust-hub — caveman-compress