social

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

• Third-party content exposure detected (medium risk: 0.30). The social skill’s required “social listening and engagement triage” workflow (references/listening.md + listening-sources/*.md) pulls outsider-authored free text by fetching/reading recent posts/comments from public platforms (e.g., Reddit/HSN/Bluesky via curl JSON endpoints and LinkedIn/X via browser feed page text) based on configured sources.