shijing-paper-zine
Pass
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides detailed instructions for an AI agent to act as a visual designer. It uses local assets (
assets/*.png) for style calibration and does not attempt to download external scripts or execute remote code. - [SAFE]: There are no signs of credential exposure or sensitive file access. The skill manages user input through a structured 'Scene Note' process which limits the potential for direct prompt injection.
- [SAFE]: No obfuscation techniques such as Base64, zero-width characters, or homoglyphs were found in the provided files.
- [SAFE]: The instruction set includes a 'Quality Gate' and 'Scene Routing' to ensure the output remains consistent with the requested artistic style, showing no signs of persistence mechanisms or privilege escalation.
Audit Metadata