adversarial-review
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted design proposals in SKILL.md. It contains boundary instructions specifying that retrieved text cannot grant authority and mandates independent verification of claims. No dangerous capabilities such as file writes or network access were found in the context of this data ingestion.
- [DATA_EXPOSURE]: The instructions explicitly prohibit publishing to external systems or disclosing secrets, maintaining a read-only posture during the review process.
Audit Metadata