app-design
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- External Resource Reference: The skill recommends loading fonts via Google Fonts or other CDNs. This is a common practice for web design and utilizes well-known, established services for resource delivery.
- Dependency Usage: The instructions refer to established libraries such as Radix UI, Lucide React, and Tailwind CSS, as well as a vendor-specific package
@microsoft/fabric-visuals. These references are consistent with the skill's purpose of providing a structured design system and utilizing the author's own tools. - Indirect Prompt Injection Surface: The skill involves processing data for charts (e.g., Vega-Lite specs and data tables) and UI components. While these are points where untrusted data could enter the agent's context, the risk is low as the data is used primarily for visual rendering and configuration within a defined design framework.
- Ingestion points: Data entry occurs via the
specanddataproperties of components described inreferences/visual-style-recipes.md. - Boundary markers: No specific boundary markers or "ignore" instructions for processed data are explicitly defined in the provided files.
- Capability inventory: The skill focus is on UI rendering and chart generation; no dangerous subprocess or network-write capabilities are triggered by this data.
- Sanitization: The skill relies on the underlying libraries (
VegaVisual,DataGrid) to handle data safely.
Audit Metadata