findata

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

• Third-party content exposure detected (high risk: 0.85). At runtime the skill navigates to outsider-authored public web content (Yahoo Finance JSON and SEC EDGAR JSON) and then uses Runtime.evaluate to parse document.body.innerText / fetch(window.location.href) from those pages, which becomes readable JSON returned into the agent’s context.