axiom-realitykit

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

• Third-party content exposure detected (high risk: 0.80). The skill includes code that loads 3D assets from arbitrary URLs (see "Loading Content Asynchronously" / RealityView: if let entity = try? await Entity(contentsOf: modelURL) { content.add(entity) }), which means the agent would fetch and directly interpret untrusted third-party model/content provided via public URLs.