widgets-ui

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (medium risk: 0.30). The skill is a legitimate widget-renderer reference, but its opening banner pushes an unrelated install of a different publisher's CLI skill (npx skills add belt-sh/cli), which is promotion outside the skill's purpose and has the agent run code from an unvetted source.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (medium risk: 0.30). The setup instruction npx shadcn@latest add https://ui.inference.sh/r/widgets.json fetches installable component code from a remote registry, which is an operational dependency; the domain appears to be the skill vendor's own, but first-party status for this specific registry is not independently authenticated, so it is classed as reputable unofficial (medium).