Abstract

Medical LLM agents remain vulnerable to adversarial input. We present AEGIS, a three-layer inline security proxy for healthcare AI agents combining an ONNX DistilBERT classifier, a locked-down LLM auditor, and HIPAA PHI redaction, with confidence-gated escalation. Across 2,165 samples, Layer 1 achieves 99.49% accuracy and 0.25% attack success rate with 0% PHI leak rate.