compose-doctor

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and parses user-generated Compose app logs and dashboard run pages (see "Step 4 — Examine Logs" with goldsky compose logs -n <app> --json and the referenced dashboard_url), and uses those untrusted third-party outputs to drive diagnosis and remediation actions, so third-party content can influence agent behavior.