2 docs tagged with "security"

Real-time threat detection for human and agent authentication — geo-velocity, baseline drift, honeytoken canaries, DPoP token binding, and Continuous Access Evaluation.

TEE LLM proxy with 20 inspection layers. Redacts secrets, blocks prompt injection, detects exfiltration, and enforces per-agent policies before forwarding to OpenAI, Anthropic, Google (Gemini), and others.