Publications

This paper introduces a new zero-trust framework for Agentic AI, focusing on decentralized identity and detailed access control to enhance security and autonomy in multi-agent systems.

arXiv:2505.19301 (2025)

This paper proposes the Agent Name Service (ANS), a universal directory designed to enable secure and seamless discovery and interaction between different AI agents.

arXiv:2505.10609 (2025)

This research outlines a methodology for developing secure Agentic AI applications by utilizing an Agent-to-Agent (A2A) protocol for protected communication and interaction.

arXiv:2504.16902 (2025)

This paper presents a zero-trust registry approach to defend multi-agent systems from 'tool squatting,' where malicious agents impersonate legitimate tools.

arXiv:2504.19951 (2025)

This paper explores the 'Trust Fabric,' a concept for decentralized coordination and economic interaction among AI agents on the emerging agentic web.

arXiv:2507.07901 (2025)

This paper introduces DIRF, a framework designed to protect the digital identities of AI agents and govern against unauthorized cloning and impersonation.

arXiv:2508.01997 (2025)

This work aims to create a standardized method for quantitatively measuring and comparing the security of different multi-agent AI systems.

arXiv:2507.21146 (2025)

This paper discusses the practical implementation and benefits of the NANDA Index Architecture from the perspective of enterprise-level applications.

arXiv:2508.03101 (2025)

This research introduces ADA, an automated defense system where adaptive AI agents protect other AI workloads using moving target defense strategies.

arXiv:2505.23805 (2025)

This paper details the Agent Capability Negotiation and Binding Protocol (ACNBP), a protocol for agents to securely negotiate and agree upon their capabilities and commitments.

arXiv:2506.13590 (2025)

This paper proposes QSAF, a new framework aimed at mitigating cognitive degradation in Agentic AI, ensuring long-term reliability and performance.

arXiv:2507.15330 (2025)