je-review

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

• Third-party content exposure detected (medium risk: 0.30). Source_JEs/JE listing input is directly ingested at runtime by scripts/validate_je.py and scripts/risk_patterns.py (via jes DataFrame loaded from the provided CSV/XLSX/tab), so outsider-authored free text fields like Memo and Support Reference are read without selecting a specific trusted item first.