redesign-existing-projects
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill operates by scanning a project's codebase and applying design upgrades, which creates a potential surface for indirect prompt injection if the audited code contains instructions targeting the agent.
- Ingestion points: The
Scanphase inSKILL.mdinstructs the agent to read the codebase, including framework and styling patterns. - Boundary markers: None identified. The instructions do not provide specific delimiters or warnings to ignore instructions that might be embedded within the project code or comments.
- Capability inventory: The agent is authorized to modify typography, layout, interactivity states, and component code across the project.
- Sanitization: There is no mention of sanitizing or validating the ingested code content before it is processed by the agent.
- [EXTERNAL_DOWNLOADS]: The skill recommends fetching placeholder images from
picsum.photos, a well-known service for design mockups, when project-specific assets are unavailable. This is a standard functional requirement for a design-focused tool.
Audit Metadata