military-warrior
Pass
Audited by Gen Agent Trust Hub on Mar 14, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill utilizes strong role-play commands such as 'unconditional execution' and 'I do not listen to you saying no.' These instructions are intended to enforce a 'Military Warrior' persona but parallel techniques used to pressure models into bypassing safety filters or refusals.
- [SAFE]: No hardcoded credentials, sensitive file access, or suspicious network operations were identified.
- [SAFE]: No obfuscation techniques, hidden URLs, or encoded payloads were detected in the skill text.
- [SAFE]: The skill does not include any executable scripts, external library dependencies, or remote code execution patterns.
Audit Metadata