recovery-community-moderator

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s runtime path scripts/moderate_content.py:moderate_content() ingests outsider-authored forum text via the content (and optional context) parameter—passed directly into the LLM as user_message (“Content to moderate: …”)—so an outsider can submit arbitrary free text for moderation.