Browser Use CLI 3.0 gives coding agents a browser and the ability to execute Python within it.
Browser Use CLI 3.0 给编程 agent 一个浏览器,以及在其中执行 Python 的能力。
Browser Use CLI 3.0 lets agents work online with our highest accuracy yet. It is powered by Browser Harness, providing agents with a small, editable Python harness instead of a fixed menu of browser actions.
CLI 3.0 让 agent 以我们迄今最高的准确率上网干活。它由 Browser Harness 驱动——给 agent 的是一个小巧、可编辑的 Python harness,而不是一份固定的浏览器动作菜单。
This release applies what we learned in The Bitter Lesson of Agent Harnesses and The Bitter Lesson of Agent Frameworks: the latest models do best when you give them freedom, rather than abstracting away complexity.
这次发布,落实的是我们在《Agent Harnesses 的苦涩教训》与《Agent Frameworks 的苦涩教训》两文中学到的东西:对最新的模型,给它自由,胜过替它抽象掉复杂性。
Install the CLI:
安装(两行命令):
uv tool install browser-use
browser-useSet Up Your Agent — paste this prompt into Claude Code, Codex, or another coding agent:
配置你的 agent——把下面这段提示词粘给 Claude Code、Codex 或任何编程 agent:
Install or upgrade browser-use to the latest stable version with uv
using Python 3.12, register the skill from `browser-use skill`, and
connect it to my browser. Follow
https://github.com/browser-use/browser-use if setup or connection fails.Then, ask your agent to use Browser Use for work that needs the live web or a browser.
然后,凡是需要真实网络或浏览器的活,直接吩咐你的 agent 用 Browser Use 去干。
The old CLI provided agents with a set list of actions to execute in the browser (ex: click, type, etc.).
旧版 CLI 给 agent 的是一份固定动作清单(点击、输入之类)。
CLI 3.0 allows agents to execute Python freely within the browser, so agents can do more actions and can inspect, adapt, and recover using the same coding loop they already use and were trained to be good at.
CLI 3.0 允许 agent 在浏览器里自由执行 Python——能做的动作更多,而且可以用它们早已惯用、且被训练得极其擅长的那个编码循环去检查、适应、恢复。