构建可信任的AI资源身份体系，实现跨框架安全发现与调用。
OpenAgenet / OAN Yellow Paper: Technical Architecture for Trust-Governed Resource Identity and Discovery
- 基于did:oan的统一身份标识，支持多协议异构代理互联
- 通过根验证包模型与凭证实现资源可信授权与访问控制
- 适合需要跨系统安全协作的AI代理平台与工具链开发者
本文描述了OpenAgenet / OAN的技术架构。OAN是一个协议无关的可信层，用于开放智能体互联与可发现的AI资源产品。它定义了角色架构、did:oan身份对象、注册流程、治理背书的根生命周期强制机制、根验证的包模型、授权感知的发现机制、根颁发的基础设施授权凭证（VCs）、签名可信调用、验证要求、状态转换、安全属性、实现边界与部署考量。该设计旨在支持异构智能体框架与交互协议，包括MCP、A2A、ANP类系统、领域特定智能体协议、技能、MCP服务器及工具/API资源。OAN不定义智能体间的完整业务对话或每种资源的原生协议；而是定义资源身份如何在协议特定交互前成为可接纳、可发现、可验证且安全可访问的。
原文摘要 · Abstract (English)
This yellow paper describes the technical architecture of OpenAgenet / OAN. OAN is a protocol-neutral trust layer for open Agent interconnection and discoverable AI resource products. It specifies the role architecture, \texttt{did:oan} identity objects, registration workflow, governance-backed Root lifecycle enforcement, Root-verified package model, authorization-aware Discovery, Root-issued infrastructure authorization VCs, signed trusted invocation, verification requirements, state transitions, security properties, implementation boundaries, and deployment considerations. The design is intended to support heterogeneous Agent frameworks and interaction protocols, including MCP, A2A, ANP-like systems, domain-specific Agent protocols, Skills, MCP Servers, and Tool/API resources. OAN does not define the entire business conversation among Agents or the native protocol of every resource; it defines how resource identities become admissible, discoverable, verifiable, and safe to approach before protocol-specific interaction begins.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。