meta-social-listening

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md workflow explicitly requires monitoring and reading public user-generated content from open sources — e.g., Google Alerts (news, blogs, forums), native platform searches on Facebook/Instagram/TikTok/X, Brand24/Mention dashboards, Google Business Profile reviews, Facebook Groups and RSS/Feedly — and instructs the agent to interpret those mentions and take actions (reply, escalate, adjust campaigns), which clearly exposes the agent to untrusted third-party content that could carry indirect instructions.