linkbuilding

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

• Third-party content exposure detected (medium risk: 0.30). Step 1/2 explicitly instructs the agent to fetch arbitrary first-party user-provided pages like “the homepage of the domain” and to use Google search/queries, which would make the model ingest outsider-authored free text from attacker-controlled websites via the provided URL/domain.