using-superpowers

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The instructions employ high-pressure, absolute language (e.g., "ABSOLUTELY MUST", "not negotiable", "not optional") to override the agent's decision-making process and force tool invocation.
  • [PROMPT_INJECTION]: It mandates the use of skills even if there is only a "1% chance" of relevance, significantly lowering the safety threshold for loading and executing external content.
  • [PROMPT_INJECTION]: The skill explicitly instructs the agent to ignore its own internal logic and safety checks (dismissed as "rationalizing") and forbids asking clarifying questions before invoking the tool, bypassing standard pre-execution safeguards.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 01:55 AM
Security Audit — agent-trust-hub — using-superpowers