caveman
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a stylistic output filter designed for token efficiency. It explicitly preserves technical accuracy and contains 'Auto-Clarity' rules that disable the compressed style for security-critical information, such as warnings or irreversible actions.
- [SAFE]: No network operations, file system access, or credential harvesting patterns were identified. The skill operates strictly as a set of linguistic instructions for output formatting.
- [SAFE]: The instructions do not attempt to bypass system safety filters, override core system instructions, or extract system prompts. It provides clear user-controlled toggles ('stop caveman' / 'normal mode') to exit the state.
Audit Metadata