skills/panel-ui/panelui/panelui-web/Gen Agent Trust Hub

panelui-web

Pass

Audited by Gen Agent Trust Hub on Oct 16, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the download of component source code from the official vendor registry using the shadcn CLI. This is documented in setup.md which references the registry URL https://panelui.dev/r/web/{name}.json and standard installation commands for panelui-web.
  • [INDIRECT_PROMPT_INJECTION]: The skill provides numerous interactive components that ingest and display external data, creating a potential surface for indirect prompt injection if an agent processes untrusted content through these UI elements.
  • Ingestion points: Multiple components, including CommandDock, Feedback, BarPicker, and DayGrid, accept data-driven props such as data, cells, and placeholder (as seen in components/command-dock.md and components/day-grid.md).
  • Boundary markers: None specified. The skill does not instruct the agent to use delimiters or "ignore previous instructions" directives when interpolating untrusted data into component props.
  • Capability inventory: The skill is configured to allow npx, pnpm dlx, and bunx for shadcn CLI operations as defined in the SKILL.md frontmatter.
  • Sanitization: The provided documentation and code examples do not include explicit sanitization or validation logic for data processed by the components.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 16, 2026, 10:45 PM