marketing-council

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

• Third-party content exposure detected (low risk: 0.10). At runtime the agent consumes user-authored free text from the question/prompt and may run a live research pass, but it never ingests outsider-authored content without explicit fetch/search actions and the required workflow primarily applies first-party advisor dossiers and the user’s provided case details.