Designing Agent-Ready Websites for AI Web Agents: A Framework for Machine Readability, Actionability, and Decision Reliability

为 AI 网络智能体设计“智能体就绪”网站：一个关于机器可读性、可操作性和决策可靠性的框架

Abstract: Online shopping is increasingly shifting toward a model in which AI agents independently search for products, compare options, evaluate constraints, and carry out parts of the purchasing process for users. Website design must now support both human and agent-mediated interaction. 摘要： 在线购物正日益转向一种新模式，即 AI 智能体能够独立为用户搜索产品、比较选项、评估约束条件并执行部分购买流程。因此，网站设计现在必须同时支持人类交互和智能体中介交互。

This paper introduces the agent-ready website, a design framework for enhancing the readability, interpretability, verifiability, and actionability of e-commerce platforms for AI agents. Existing web design, SEO, and generative engine optimization (GEO) metrics do not fully assess a website’s capacity for agent-mediated interaction. 本文引入了“智能体就绪网站”（agent-ready website）的概念，这是一个旨在增强电子商务平台对 AI 智能体的可读性、可解释性、可验证性和可操作性的设计框架。现有的网页设计、SEO（搜索引擎优化）以及生成式引擎优化（GEO）指标，尚无法全面评估网站对智能体中介交互的承载能力。

The proposed framework is structured around three dimensions—agent interpretability, agent executability, and agent decision reliability—supported by features such as machine readability, semantic clarity, agent actionability, and contextual decision-reliability signals. 该框架围绕三个维度构建：智能体可解释性、智能体可执行性和智能体决策可靠性。这些维度由机器可读性、语义清晰度、智能体可操作性以及上下文决策可靠性信号等功能特性提供支持。

The framework is evaluated through a controlled experiment comparing a human-oriented baseline and an agent-ready version of an identical website prototype, with identical catalogs, pricing, stock, and shopping workflows. The evaluation involved five tasks, three browser-agent models (GPT-4.1, Gemini-2.5 Flash, and Grok-4 Fast), and 300 runs, measuring PASS, PARTIAL, FAIL outcomes, strict and functional success rates, error patterns, step counts, and token consumption. 该框架通过一项对照实验进行了评估，对比了以人类为导向的基准网站与“智能体就绪”版本的相同网站原型，两者具有完全一致的目录、定价、库存和购物流程。评估涵盖了五项任务、三种浏览器智能体模型（GPT-4.1、Gemini-2.5 Flash 和 Grok-4 Fast）以及 300 次运行，测量了通过（PASS）、部分通过（PARTIAL）、失败（FAIL）的结果，以及严格和功能性的成功率、错误模式、步骤计数和 Token 消耗量。

The agent-ready website achieved 134 PASS runs out of 150 versus 74 out of 150 for the baseline (strict success rates of 89.3% vs. 49.3%), with the largest gains in product detail extraction, comparison, and multi-constraint selection. It also reduced PARTIAL outcomes from 43 to 3 and lowered the average step count from 9.31 to 6.49. “智能体就绪”网站在 150 次运行中取得了 134 次通过，而基准网站仅为 74 次（严格成功率分别为 89.3% 和 49.3%），其中在产品详情提取、比较和多约束选择方面的提升最为显著。此外，该框架将“部分通过”的结果从 43 次减少到 3 次，并将平均步骤计数从 9.31 步降低至 6.49 步。

These results provide preliminary evidence that enhanced structural clarity, action cues, evidence signals, and temporal validity indicators can substantially improve the reliability and efficiency of AI browser agents. 这些结果提供了初步证据，表明增强结构清晰度、操作提示、证据信号和时间有效性指标，可以显著提高 AI 浏览器智能体的可靠性和效率。