refresh-content

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

• Third-party content exposure detected (medium risk: 0.65). The required workflow calls external APIs (XAI for X and YouTube API) at runtime and then parses/validates the returned content, which can include outsider-authored free text (e.g., post/mention bodies) that is ingested into the agent’s LLM context during summarization (video-summaries) or downstream processing.