portfolio-story-architecture-by-1srj4fh
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references various external assets including videos (mp4), images (webp, gif, png), and font files. These resources are hosted on domains like goodcase.ai, motionsites.ai, cloudfront.net, figma.site, and onlinewebfonts.com. These are used as design references or incorporated as assets in the generated website code.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied information (e.g., page goals, brand context, product assets) to generate website code and narratives. While this creates a potential surface for processing untrusted data, the skill includes explicit safety instructions that prohibit the agent from making external changes or publishing without user approval.
- [COMMAND_EXECUTION]: The workflow directs the agent to use available tools, such as a browser or image tool, to inspect finished media from the reference cases. This is a standard and expected use of agent capabilities for content verification.
Audit Metadata