Agent 与自动化 4.0 · 优秀 2026-09-03 · 论文

The Natural Language Interaction Protocol and Standard for AI Agents

NLIP(自然语言交互协议):由多家公司与高校的研究者和从业者共同制定经 Ecma International 标准化的 AI 智能体应用层通信协议针对异构开发框架模型工具接口与执行环境并存的现状,NLIP 提供轻量语义消息信封,可承载于 HTTP/HTTPSWebSocketAMQP 等现有传输之上,让 NLIP 感知的智能体与网关在客户端智能体本地上下文存储本体工具与企业服务之间完成适配互通

打开原文回到归档

The Natural Language Interaction Protocol and Standard for AI Agents

  • ID: 04be3912
  • 原文链接: https://arxiv.org/abs/2609.04135
  • PDF: https://arxiv.org/pdf/2609.04135v1
  • 作者: 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
  • 日期: 2026-09-03
  • 更新: 2026-09-03
  • 分类: cs.AI
  • 备注: Accepted by ACM AI Summit 2026
  • 来源类型: arxiv
  • 标签: interop, protocol, standardization, ecma, nlip, arxiv
  • 质量评分: 4/5
  • 抓取时间: 2026-09-06T04:24:00Z

中文导读

组织内的 AI 智能体正由异构的开发框架、模型、工具接口、协议与执行环境构建,要释放社会与商业价值就必须通过共同协议互操作。NLIP(Natural Language Interaction Protocol)由跨公司与高校的研究者和从业者制定,并经 Ecma International 标准化,是面向 AI 智能体交互的基于标准的应用层协议:提供轻量语义消息信封,可承载于 HTTP/HTTPS、WebSocket、AMQP 等现有传输之上,让 NLIP 感知的智能体与网关在客户端、智能体、本地上下文存储、本体、工具与企业服务之间完成适配。论文覆盖设计动机与理由、消息模型与传输绑定、安全设计(security-by-design)考虑、参考实现、代表性应用、采用信号,以及与 MCP、A2A 等新兴智能体协议的关系。

为什么值得关注

智能体互操作进入标准时代:Ecma 标准化的 NLIP 用轻量语义信封跑在 HTTP/WebSocket/AMQP 上,对标 MCP/A2A 之外的第三个协议坐标

English Abstract

AI agents are increasingly being developed and deployed across organizations using heterogeneous agent-development frameworks, AI models, tool interfaces, protocols, and execution environments. To realize their potential social and business impact, these agents must be able to interoperate through a common communication protocol. The Natural Language Interaction Protocol (NLIP), developed by researchers and practitioners across companies and universities and standardized by Ecma International, addresses this need by defining a standards-based application-layer protocol for AI-agent interaction.

Obsidian 证据

  • 元数据与摘要经 opencli arxiv paper 2609.04135 核对(2026-09-06T04:24:00Z);中文导读锚定摘要陈述的事实与数字。