navra

Papers

Research papers covering navra's security architecture, persona orchestration, autonomous review, audit logging, and model card schema.

Always-On Audit for AI Agent Gateways: A Hash-Chained Blackbox Approach →
Persona-Driven Multi-Agent Orchestration: A General-Purpose Cognitive Framework for AI Agent Teams →
navra: A Security Microkernel for AI Agent Infrastructure →
Domain-Agnostic Autonomous Review via Dynamic Persona Selection in Multi-Agent Flows →
Composite Model Cards for Agentic AI: Bridging Registry Metadata and Runtime Selection →