community-hottopic-insight

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

• Third-party content exposure detected (medium risk: 0.30). In SKILL.md Step 1–3, the runtime pulls outsider-authored community content via ae-cli community +search_posts and then ingests full outsider post text/comments via ae-cli community +get_post_detail (used to extract quotes and sentiment), meaning attackers can post free text into the monitored community feeds for the workflow to read.