Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Agent-OSI: A Layered Protocol Stack Toward a Decentralized Internet of Agents

Wenxin Xu, Taotao Wang, Yihan Xia, Shengli Zhang, Soung Chang Liew

TL;DR

Results show that keeping negotiation and delivery off-chain while preserving verifiable settlement reduces on-chain session costs by approximately 51\% compared with a standard Web3 baseline in the prototype setting, and that blockchain confirmation latency is often not the dominant factor for generative workloads.

Abstract

Large Language Models (LLMs) are accelerating the shift from an Internet of information to an Internet of Agents (IoA), where autonomous entities discover services, negotiate, execute tasks, and exchange value. Yet today's agents are still confined to platform silos and proprietary interfaces, lacking a common stack for interoperability, trust, and pay-per-use settlement. This article proposes \textit{Agent-OSI}, a six-layer reference stack for decentralized agent networking built on top of the existing Internet. Agent-OSI combines secure connectivity and A2A messaging, decentralized identity and authorization, settlement and metering, verifiable execution and provenance, and semantic interoperability for orchestration. In particular, we treat HTTP 402 (Payment Required) as an application-level payment challenge (analogous to HTTP 401 for authentication) that triggers escrow-based settlement and verifiable receipts (instantiated via a blockchain escrow in our prototype), rather than introducing a new network-layer protocol. We implement a prototype and evaluate cost and latency. Results show that keeping negotiation and delivery off-chain while preserving verifiable settlement reduces on-chain session costs by approximately 51\% compared with a standard Web3 baseline in our prototype setting, and that blockchain confirmation latency is often not the dominant factor for generative workloads.

Agent-OSI: A Layered Protocol Stack Toward a Decentralized Internet of Agents

TL;DR

Results show that keeping negotiation and delivery off-chain while preserving verifiable settlement reduces on-chain session costs by approximately 51\% compared with a standard Web3 baseline in the prototype setting, and that blockchain confirmation latency is often not the dominant factor for generative workloads.

Abstract

Large Language Models (LLMs) are accelerating the shift from an Internet of information to an Internet of Agents (IoA), where autonomous entities discover services, negotiate, execute tasks, and exchange value. Yet today's agents are still confined to platform silos and proprietary interfaces, lacking a common stack for interoperability, trust, and pay-per-use settlement. This article proposes \textit{Agent-OSI}, a six-layer reference stack for decentralized agent networking built on top of the existing Internet. Agent-OSI combines secure connectivity and A2A messaging, decentralized identity and authorization, settlement and metering, verifiable execution and provenance, and semantic interoperability for orchestration. In particular, we treat HTTP 402 (Payment Required) as an application-level payment challenge (analogous to HTTP 401 for authentication) that triggers escrow-based settlement and verifiable receipts (instantiated via a blockchain escrow in our prototype), rather than introducing a new network-layer protocol. We implement a prototype and evaluate cost and latency. Results show that keeping negotiation and delivery off-chain while preserving verifiable settlement reduces on-chain session costs by approximately 51\% compared with a standard Web3 baseline in our prototype setting, and that blockchain confirmation latency is often not the dominant factor for generative workloads.
Paper Structure (32 sections, 3 figures, 2 tables)

This paper contains 32 sections, 3 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Why Decentralized Agent Networks?
  3. Interoperability: Beyond Platform Silos
  4. Sovereignty: Portable Identity and Asset Ownership
  5. Trust: Adversarial and Cross-Domain Settings
  6. Deployability: Incremental Adoption on Today's Internet
  7. The Agent-OSI Architecture
  8. L1: Connectivity and Secure Transport
  9. L2: A2A Messaging and Routing
  10. L3: Identity, Authentication, and Authorization
  11. L4: Settlement and Metering
  12. 402 Payment Challenge Message
  13. Receipt and Verification
  14. L5: Verifiable Execution and Provenance
  15. L6: Semantic Interoperability and Orchestration
  16. ...and 17 more sections

Figures (3)

  • Figure 1: The Agent-OSI reference stack, a six-layer agent stack built on top of today's Internet stack. While layering guides modularity, some end-to-end properties are enforced via well-defined signed artifacts across layers (detailed in Section \ref{['sec:agent_osi']}).
  • Figure 2: Sequence diagram of the cross-layer interaction. The workflow coordinates A2A messaging, payment challenge and settlement, verifiable execution evidence, and delivery.
  • Figure 3: Performance of the Agent-OSI prototype. (a) Cost: on-chain EVM execution cost per completed paid session (sum of gasUsed; network fees excluded); Web 2.0 baseline (off-chain billing) has zero on-chain cost and is omitted. (b) Latency: end-to-end breakdown (messaging/network, confirmation, execution/delivery) for Light (no-gen), Pipeline ($K$-step, IPFS/CID), and GenAI (image/LLM, IPFS/CID). (c) Throughput: steady-state message (L2), settlement (L4 confirmed escrow tx), and completed-session throughput under increasing concurrency (UAs$\rightarrow$1 SA).