ads

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

• Third-party content exposure detected (high risk: 1.00). The skill's Execution Steps (Step 1) explicitly fetch and analyze provided landing-page URLs — "If cache doesn't exist yet, fetch the URL, analyze it, and save cache" and "If a raw URL was provided ... fetch and analyze it directly" — so arbitrary public web content would be read and used to drive ad-copy generation and subsequent actions.