integrated-browser

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

• Third-party content exposure detected (high risk: 0.75). Outsider free text can enter the LLM context via the integrated browser’s agent tooling when it loads a user-supplied web page (public/remote page body text) and then the agent reads/summarizes that page content through the browser tools (screenshots/DOM-derived text become model input).