quality-score-fixer

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow ingests user-provided Google Ads data (e.g., “paste keyword metrics…current ads, and landing-page copy” and then runs a READ-ONLY diagnosis based on that pasted/connected content), so outsider-authored free text entered by a user can be directly consumed by the agent.