tiktok

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

• Third-party content exposure detected (high risk: 0.75). The required workflow is to call TikHub API endpoints (e.g., fetch video/user/comments/search results) and pass the returned response text into downstream task skills/LLM context; since the content is sourced from an external service (public TikTok/TikHub data), it is outsider-authored free text at runtime.