troubleshoot

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

• Third-party content exposure detected (high risk: 0.90). The "Frontend Verify" and "Authentication for Frontend Verify" sections explicitly instruct using agent-browser to open arbitrary login URLs, navigate SaaS/admin pages, and automate interactions (open , snapshot/fill/click, state save/load), so the agent will fetch and act upon untrusted public web content as part of its workflow.