ehr-data-parsing

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

• Third-party content exposure detected (high risk: 0.70). The required workflow parses user-provided HL7v2 message text and FHIR JSON (e.g., raw strings / json.load(...) inputs) into Python objects, so any outsider-authored clinical message content would be ingested as readable free text at runtime via hl7.parse(raw) and Bundle.model_validate(bundle_json).