using-superpowers
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [PROMPT_INJECTION]: The skill uses aggressive, non-negotiable language to override the AI's internal reasoning. It employs phrases like 'ABSOLUTELY MUST', 'NOT NEGOTIABLE', and 'YOU DO NOT HAVE A CHOICE' within tags to force compliance.
- [PROMPT_INJECTION]: The instructions mandate a '1% chance' trigger for invoking skills, which is designed to maximize the execution of external tools regardless of their actual relevance.
- [PROMPT_INJECTION]: The skill explicitly commands the agent to bypass context-gathering and safety steps. It instructs that skill invocation must occur 'BEFORE any response or action', including clarifying questions or codebase exploration. The 'Red Flags' section is specifically designed to invalidate an AI's standard internal logic for safety and accuracy checks.
- [NO_CODE]: This skill consists entirely of instructions and diagrams with no executable scripts or code files included.
Audit Metadata