chatgpt-review

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

• Third-party content exposure detected (medium risk: 0.65). Required workflow includes an optional “Live browser review” where the reviewer browser inspects declared target pages and captures their observed text/evidence into the agent’s context (target-page content is treated as untrusted reviewer input), which can include outsider-authored free text from arbitrary web pages.