clay

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly fetches and ingests public, user-generated content (see references/api-integration.md — e.g., SketchfabSource, ObjaverseSource, PolyPizzaSource and the "External Model Download" / search_all_sources and download_with_validation functions, plus SKILL.md Output Routing entries for "download/fetch model"), and those downloads and metadata are consumed as part of the agent's workflow and downstream decision-making, so untrusted third-party content could materially influence actions.