instagram-audience-voice

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly runs hosted collection actors (primary collection key "instagram-comments" and the instagram-comment-scraper) to fetch and normalize public Instagram post/reel comments (see SKILL.md "collect comments from a shortlisted set of posts or reels" and the actor/tool contracts), which are untrusted, user-generated third-party content that the agent reads and uses to cluster/summarize and drive downstream decisions like audience-language and copy selection.