skills/fusengine/agents/react-shadcn/Gen Agent Trust Hub

react-shadcn

Pass

Audited by Gen Agent Trust Hub on Feb 28, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADS
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the installation of component source code using the command bunx shadcn@latest add. This process fetches code from the official shadcn/ui registry, which is a trusted and established service in the React development community.
  • [EXTERNAL_DOWNLOADS]: Component examples and references include media assets and configuration schemas hosted on well-known, trusted domains such as images.unsplash.com, github.com, and ui.shadcn.com.
  • [PROMPT_INJECTION]: The components defined in the skill (such as Data Tables, Forms, and Breadcrumbs) act as ingestion points for external data, creating a surface for potential indirect prompt injection.
  • Ingestion points: Identified in references/table.md (table rows), references/form-examples.md (form fields), and references/breadcrumb.md (URL path segments).
  • Boundary markers: Absent; components render data directly as strings or numbers within the UI structure.
  • Capability inventory: The skill includes scripts that perform network requests via fetch and system-level operations via the bunx command for component generation.
  • Sanitization: The components rely on React's built-in string escaping for protection against XSS, but they do not implement specific logic to sanitize or ignore instructions embedded in the data being rendered.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 28, 2026, 11:02 AM