retro-dv-home-video-by-6moylp
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No security issues detected. The skill contains instructions that explicitly forbid making external account changes without user approval and emphasizes proper attribution.
- [DATA_EXPOSURE_AND_EXFILTRATION]: The skill references external URLs for video and image media on goodcase.ai and x.com. These are utilized for visual style analysis and do not involve harvesting sensitive system or user information.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external prompt excerpts and media summaries. It manages this risk through a mandatory reference contract that defines what to preserve, replace, or avoid, ensuring the agent remains focused on the user's intent while incorporating stylistic elements.
Audit Metadata