anysite-content-analytics

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly fetches and analyzes user-generated social media content (e.g., get_twitter_user_posts, get_youtube_video_comments, reddit_user_posts, get_instagram_user_posts) from public platforms, so the agent will read untrusted third-party posts/comments that could contain indirect prompt injections.