skills/rafaelgrasa/skills/widgets-ui/Gen Agent Trust Hub

widgets-ui

Pass

Audited by Gen Agent Trust Hub on Apr 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill utilizes the shadcn CLI to fetch and install UI component definitions from https://ui.inference.sh/r/widgets.json. This is the official registry for the vendor's UI framework.
  • [COMMAND_EXECUTION]: Installation instructions include running npx skills add commands to fetch and integrate additional agent-related components from the inference-sh skill registry.
  • [PROMPT_INJECTION]: The skill implements a Generative UI pattern which serves as a surface for indirect prompt injection. 1. Ingestion points: The WidgetRenderer component accepts a widget JSON object directly from agent-generated output (SKILL.md). 2. Boundary markers: The documentation does not specify the use of delimiters or schemas to enforce boundaries between UI structural elements and potentially malicious agent-provided content. 3. Capability inventory: The system supports interactive elements including buttons, forms, and input fields that can trigger callbacks or capture user data (SKILL.md). 4. Sanitization: No specific sanitization or validation mechanisms are described for the JSON payload before it is rendered into an active UI.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 14, 2026, 11:26 PM