google-safe-browsing

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

• Third-party content exposure detected (high risk: 0.70). The required workflow uses curl to fetch a potentially arbitrary flagged website page body at runtime (Step “Fetch the site anonymously (curl -sL), see exactly what Googlebot sees”), which can include outsider-authored free text that becomes LLM-readable context via the tool/output.