ad-research

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly scrapes and ingests public competitor ads and landing pages (e.g., Meta Ad Library at facebook.com/ads/library via Playwright and web searches) and uses their copy/transcripts to drive analysis and downstream ad-generation decisions, which exposes the agent to untrusted third-party content that could enable indirect prompt injection.