velt-comments-best-practices

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 1.00). The skill instructs the agent to ingest and act on user-generated comment annotations fetched from the Velt service (e.g., via useCommentAnnotations, client.getCommentElement(), VeltCommentsSidebar handlers and VeltProvider/VeltComments examples) and explicitly uses those annotations' context/location to drive actions like navigation or seeking, so untrusted third-party content can influence behavior.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill includes an HTML runtime loader that fetches and executes remote JavaScript from https://cdn.velt.dev/lib/sdk@latest/velt.js (via a ), which is a required SDK dependency and thus executes remote code at runtime.