designing-uis

Pass

Audited by Gen Agent Trust Hub on Sep 29, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to browse external design portfolios and GitHub repositories to extract design patterns. This research process exposes the agent to potentially untrusted content that could contain instructions designed to influence its behavior.\n
  • Ingestion points: Content from external websites like Dribbble and designer-specific repositories on GitHub.\n
  • Boundary markers: There are no instructions to use specific delimiters or to treat the retrieved text as data rather than instructions.\n
  • Capability inventory: The skill utilizes file writing capabilities for creating UI specifications and mockups, and employs browser tools for site navigation.\n
  • Sanitization: The instructions do not specify any validation or filtering for the data retrieved during the research phase.\n- [EXTERNAL_DOWNLOADS]: The provided mockup files include links to external libraries and assets, such as Google Fonts and the Basecoat and Lucide libraries hosted on the JSDelivr CDN. It also directs the agent to research design portfolios on Dribbble and specific design templates on GitHub.\n
  • Evidence: Fetches resources from fonts.googleapis.com, cdn.jsdelivr.net, dribbble.com, and github.com/davidodunjo/design-mds.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 29, 2026, 08:58 AM