Beyond DNS: Unlocking the Internet of AI Agents via the NANDA Index and Verified AgentFacts
🤖 AI Summary
Existing DNS mechanisms are inadequate for a trillion-scale autonomous AI agent internet, failing to support millisecond-scale negotiation, dynamic agent migration, and fine-grained permission management. To address this, we propose NANDA, a novel indexing architecture featuring a hypertree-based distributed index that unifies discoverability for both native and third-party agents. NANDA introduces sub-second cryptographic key rotation and revocation, capability assertions validated via schema-based attestation, and CRDT-driven distributed state synchronization. It further defines a cryptographically verifiable query protocol enabling minimal disclosure of sensitive information. Evaluated in a prototype compatible with existing web infrastructure, NANDA achieves global sub-second resolution, cross-domain privacy-preserving discovery, dynamic load balancing, and trusted capability exchange. The system demonstrates horizontal scalability and strong security guarantees—including confidentiality, integrity, and availability—under adversarial conditions.
Application Category
📝 Abstract
The Internet is poised to host billions to trillions of autonomous AI agents that negotiate, delegate, and migrate in milliseconds and workloads that will strain DNS-centred identity and discovery. In this paper, we describe the NANDA index architecture, which we envision as a means for discoverability, identifiability and authentication in the internet of AI agents. We present an architecture where a minimal lean index resolves to dynamic, cryptographically verifiable AgentFacts that supports multi-endpoint routing, load balancing, privacy-preserving access, and credentialed capability assertions. Our architecture design delivers five concrete guarantees: (1) A quilt-like index proposal that supports both NANDA-native agents as well as third party agents being discoverable via the index, (2) rapid global resolution for newly spawned AI agents, (3) sub-second revocation and key rotation, (4) schema-validated capability assertions, and (5) privacy-preserving discovery across organisational boundaries via verifiable, least-disclosure queries. We formalize the AgentFacts schema, specify a CRDT-based update protocol, and prototype adaptive resolvers. The result is a lightweight, horizontally scalable foundation that unlocks secure, trust-aware collaboration for the next generation of the Internet of AI agents, without abandoning existing web infrastructure.
Problem
Research questions and friction points this paper is trying to address.
Enabling discoverability and authentication for AI agents
Replacing DNS for dynamic AI agent identity management
Ensuring privacy and scalability in AI agent collaboration
Innovation
Methods, ideas, or system contributions that make the work stand out.
NANDA index enables AI agent discovery
Cryptographically verifiable AgentFacts ensure security
CRDT-based protocol supports dynamic updates
🔎 Similar Papers
No similar papers found.
