marketing-influencer

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

• Third-party content exposure detected (high risk: 1.00). This skill's Research Mode (Influencer Discovery) explicitly uses agent-browser commands to open public social platforms (e.g., "https://www.tiktok.com/...", "https://www.instagram.com/...", "https://www.youtube.com/...") and "agent-browser get text body" to ingest and evaluate user-generated third-party content for vetting and decision-making, which clearly exposes the agent to untrusted web content that can influence its actions.