competitor-ad-intelligence

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

• Third-party content exposure detected (medium risk: 0.30). SKILL.md Phase 1–4 ingests outsider-authored free text by using web_search and fetch_webpage to scrape competitor ads from Meta Ad Library and Google Ads Transparency Center (and then fetches landing pages discovered in those ads), which can include attacker-controlled poison content submitted by any third party.