aile-using-superpowers
Pass
Audited by Gen Agent Trust Hub on Feb 28, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [PROMPT_INJECTION]: The skill uses authoritative language like "MUST," "NO CHOICE," and "Hard rule" to override the agent's autonomy and force a specific tool-based workflow.- [PROMPT_INJECTION]: The "Danger Signals" section explicitly instructs the agent to disregard standard reasoning steps—such as seeking context or asking clarifying questions—in favor of immediate tool invocation.- [PROMPT_INJECTION]: It establishes a "1% probability" threshold for tool invocation, which serves to bypass the agent's internal relevance and safety assessments when deciding to load and execute external skills.
Audit Metadata