using-superpowers

Pass

Audited by Gen Agent Trust Hub on Mar 26, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill uses high-pressure, absolute language ("ABSOLUTELY MUST", "not negotiable", "not optional") designed to override the agent's standard decision-making processes.
  • [PROMPT_INJECTION]: It explicitly instructs the agent to disregard its own reasoning or judgment, characterizing typical agent behaviors (like seeking context or exploring codebases) as 'rationalizing' and 'red flags'.
  • [PROMPT_INJECTION]: The instructions mandate that the agent must invoke tool-checking logic before providing any response to the user, effectively bypassing normal interaction flows to ensure skill-based instructions are always prioritized.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 26, 2026, 04:43 PM