Voice of Customer Summarizer

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md explicitly requires ingesting and analyzing untrusted, user-generated content (e.g., social_mentions from Twitter/X, Instagram, Facebook, Reddit, TikTok, community_posts, and marketplace product_reviews) and uses that content to drive theme extraction, viral-risk assessments, and recommended actions, so third-party content can materially influence behavior.