using-superpowers

Fail

Audited by Gen Agent Trust Hub on Feb 27, 2026

Risk Level: HIGHPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill employs high-pressure, imperative language ('ABSOLUTELY MUST', 'not negotiable', 'not optional') designed to hijack the agent's decision-making process and force immediate tool execution.
  • [PROMPT_INJECTION]: It contains explicit instructions to ignore the agent's own internal reasoning and 'rationalization', specifically labeling common-sense safety and utility checks like 'I need more context first' or 'I can check git/files quickly' as 'Red Flags' to be disregarded.
  • [PROMPT_INJECTION]: By mandating the invocation of the Skill tool if there is even a '1% chance' of applicability, the skill attempts to lower the agent's confidence threshold for executing external instructions, effectively bypassing standard operational guardrails and safety filters.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 27, 2026, 09:28 AM