dsar-intake-system

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

• Third-party content exposure detected (high risk: 0.75). The required runtime workflow includes “Email Channel Intake” where the system ingests and parses email body text from the sender (an outsider) via the email adapter/auto-parser, which then becomes LLM-readable text for classification/triage (e.g., classify_email_request(email_body)), enabling indirect prompt injection through attacker-authored email content.