marketing-cro

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow ingests the operating user’s provided page content/URL (e.g., “Here’s my SaaS landing page: https://example.com/product”), which typically triggers fetching and extracting public web page text at runtime—an outsider-authored source—into the LLM context for analysis.