AI Agents的自然语言交互协议与标准

Luyi Xing, Rasit Onur Topaloglu, Ranjan Sinha, Abhay Ratnaparkhi, Samuel Ndichu, Christopher Nguyen, Anindita Das, Tom Sheffler, Mohamed Rahouti, Zichuan Li, Xiaojing Liao, Sanjay Aiyagari

论文 2026 原文 2026-09-03 arXiv preprint arXiv:2609.04135 收录 2026-10-04 protocolstandardnatural-languageinteroperability

摘要

AI agents正基于异构框架、模型、工具接口和协议被广泛开发部署，需通过通用通信协议实现互操作。由多机构研究者开发、Ecma International标准化的NLIP，定义了基于标准的AI agent交互应用层协议。它提供轻量级语义消息封装，可承载于HTTP/HTTPS、WebSocket和AMQP等现有传输之上，并允许感知NLIP的agent和网关在客户端、agent、本地上下文存储、本体、工具、企业服务及异构底层协议间适配。本文介绍NLIP的动机与设计原理、消息模型与传输绑定、安全设计、参考实现、典型应用、采用信号及其与MCP、A2A等新兴agent协议的关系。

引用本文条目

GB/T 7714-2015

Luyi Xing, Rasit Onur Topaloglu, Ranjan Sinha, et al. The Natural Language Interaction Protocol and Standard for AI Agents[EB/OL]. arXiv preprint, 2026(2026-09-03)[2026-10-06]. https://arxiv.org/abs/2609.04135.

BibTeX

@misc{xing2026,
  author = {Luyi Xing and Rasit Onur Topaloglu and Ranjan Sinha and Abhay Ratnaparkhi and Samuel Ndichu and Christopher Nguyen and Anindita Das and Tom Sheffler and Mohamed Rahouti and Zichuan Li and Xiaojing Liao and Sanjay Aiyagari},
  title = {The Natural Language Interaction Protocol and Standard for AI Agents},
  year = {2026},
  organization = {arXiv preprint},
  howpublished = {\url{https://arxiv.org/abs/2609.04135}},
}

查看原文 ↗

本条目为「智联观察」资料库收录，引用时请注明来源与本页链接。