instagram-audience-voice

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly collects and processes public, user-generated Instagram content (e.g., "collect comments from a shortlisted set of posts or reels" in SKILL.md and the use of actors like instagram-comment-scraper in actor-selection.md and the runHostedCollection/run_instagram_actor scripts), so the agent will ingest untrusted third-party comments that can influence analysis and downstream actions.