ad-creative-intelligence

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

• Third-party content exposure detected (high risk: 0.85). At runtime the workflow scrapes outsider-authored public web pages (Meta Ads Library and Google Ads Transparency) via the provided URLs, converts them to readable text (--body {"format":"text"}), and then passes that scraped ad text into the LLM for analysis (Step 4), which is indirect prompt-injection exposure.