跳到正文
arXiv cs.AI· Dario Fenoglio, Martin Gjoreski, Marc Langheinrich, Pradyumna Chari, Ramesh Raskar·· 56 分钟前AI 评分39

WebAgents:为智能体原生 Web 解耦智能体的创建、托管与执行

WebAgents: Unbundling Agent Creation, Hosting, and Execution for an Agent-Native Web

AI 导读

WebAgents 通过签名的可移植包,将智能体的创建、托管与执行角色解耦,使智能体工作流可在不同创建者、主机与运行时之间迁移。基于 OpenClaw 的智能体原生浏览器/运行时会验证包、获取授权,并运行时绑定用户控制的模型、工具、权限与预算。

正文 · 原文

View PDF HTML (experimental)

Abstract:Large language models are moving from isolated assistants toward autonomous agents that can plan, use tools, and coordinate with others, motivating an emerging Agentic Web of independently developed agents and services. Realizing this vision requires agent behavior to move across creators, hosts, and runtimes without coupling publication to a particular execution environment or transferring authority over user resources. We introduce WebAgents, which unbundles these roles through a signed this http URL that carries portable workflows, skills, schemas, and resource requirements. An OpenClaw-based agent-native browser/runtime verifies the package, obtains authorization, and binds it at run time to user-controlled models, tools, permissions, and budgets, while hosts distribute the public program and may retain private services server-side. We demonstrate this contract with three unrelated workflows executed unchanged under local and API-backed models, including literature review with live services and a website-published travel agent using semantic tools and user-selected personal resources. In a controlled suite, WebAgents explicitly blocks all 716 injected contract violations while accepting all 295 matched benign controls, with median automated pre-execution processing below 0.5 s for all three packages. We conclude by outlining a path toward NANDA-based discovery and interoperable package ecosystems while retaining execution authority in the user's runtime.
Comments: Presented at IEEE CIC 2026. 8 pages
Subjects: Artificial Intelligence (cs.AI)
Cite as: arXiv:2611.21978 [cs.AI]
  (or arXiv:2611.21978v1 [cs.AI] for this version)
  https://doi.org/10.48550/arXiv.2611.21978

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Dario Fenoglio [view email]
[v1] Mon, 16 Nov 2026 23:05:11 UTC (90 KB)

来源:arXiv cs.AI · arxiv.org