anomaly-scan

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly ingests data from connected third‑party marketing platforms (Process step 2: "Pull current metrics from all connected MCPs" — e.g., google-analytics, google-ads, meta-marketing, tiktok-ads, shopify, etc.), and that external/untrusted content is read and interpreted to drive anomaly detection and recommended actions, so it can materially influence agent behavior and enable indirect prompt injection.