monitor-website

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

• Third-party content exposure detected (medium risk: 0.30). In monitor-website, the runtime path scripts/design_review.py’s preflight_result() directly fetches outsider-authored free text/markup from a user-supplied --review-url (via fetch(args.review_url)), extracts text from that response, and uses it to drive subsequent workflow decisions—without first selecting a specific trusted item.