using-superpowers
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEPROMPT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill employs highly imperative and absolute language to override the agent's autonomous reasoning. Phrases like "ABSOLUTELY MUST," "not negotiable," "not optional," and "YOU DO NOT HAVE A CHOICE" are used to force the agent to prioritize skill invocation over its own internal logic or safety assessments. It explicitly instructs the agent to disregard its own "rationalizations" for not using a skill.
- [INDIRECT_PROMPT_INJECTION]: The skill establishes a mandatory workflow for loading external content via the Skill tool based on a "1% chance" of relevance. This significantly lowers the barrier for loading potentially malicious skill files into the agent's context.
- Ingestion points: The
Skilltool is used to ingest content from external skill files. - Boundary markers: None present; the skill instructs the agent to "Follow skill exactly" rather than treating external content with skepticism.
- Capability inventory: The skill identifies usage of the
Skilltool and theTodoWritetool. - Sanitization: No sanitization or validation steps are defined for the content loaded from the Skill tool.
Audit Metadata