caveman

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a stylistic modification ('caveman mode') intended to reduce token consumption by removing filler words and articles. It does not attempt to bypass core safety guardrails.
  • [SAFE]: No malicious prompt injection, data exfiltration, or unauthorized command execution patterns were identified. The instructions are focused entirely on communication style.
  • [SAFE]: The skill includes an explicit 'Auto-Clarity Exception' that prioritizes security warnings and critical user interactions (like irreversible actions) over stylistic constraints, ensuring that safety-critical information is always delivered clearly.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 06:00 AM
Security Audit — agent-trust-hub — caveman