interview-system-designer

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

• Third-party content exposure detected (medium risk: 0.30). The workflow reads candidate interview results and role definitions from JSON input files (assets/sample_interview_results.json, assets/sample_role_definitions.json, or arbitrary --input paths) and analyzes them with hiring_calibrator.py. Because candidate evaluation data and interview feedback can be authored by external parties or applicants, this represents an indirect prompt injection risk via outsider-submitted text.