Webwright: Why AI Web Agents Should Write Code, Not Click
中文摘要
微软 Webwright 让 AI 代理通过编写代码而非点击来操作，将长任务成功率提升至 60.1% 并生成可复用的命令行工具。
English Summary
Microsoft's Webwright lets web agents write code instead of clicking, boosting long-task success rates to 60.1% and creating reusable command-line tools.
原文节选
For years, web agents have worked one click at a time—and often fallen apart on long tasks. Microsoft Research’s Webwright makes a different bet: give the model a terminal and let it write the program instead. On long-horizon tasks, the same GPT-5.4 model jumps from 33.5% to 60.1% success. And instead of leaving behind a click trace, it leaves something you can actually use again: a command-line tool. The post Webwright: Why AI Web Agents Should Write Code, Not Click appeared first on Towards Data Science.