poster-information-hierarchy-by-1e0klp2
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates as a prompt engineering framework that uses local reference files to ground AI generations in specific design styles. Analysis of the instructions and the referenced
cases.mdfile shows no evidence of malicious patterns. - [EXTERNAL_DOWNLOADS]: The skill contains several links to external websites (goodcase.ai, x.com) and media files (media.goodcase.ai). These are provided for reference and source attribution purposes and are documented as standard external resources. There is no automated execution or suspicious downloading of remote code.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a clear ingestion surface by reading
references/cases.mdand asking for user inputs. However, it implements robust boundary markers and verification steps (Preserve/Replace/Avoid contract, verification checklists) that significantly reduce the risk of accidental obedience to instructions embedded within the data. Given the design-focused nature of the output, the attack surface for sensitive operations is non-existent.
Audit Metadata