Back to directory

agent-browser

agent-browser is a fast, native Rust CLI for driving a real Chrome browser from an AI agent, using accessibility-tree snapshots and stable element references instead of fragile CSS selectors.

agent-browser

A snapshot command returns the page's accessibility tree with numbered references, and later commands click, fill or read an element by that reference; traditional CSS-selector and role-based commands are also supported for cases where that is more direct. It downloads and manages its own Chrome for Testing build, or detects an existing Chrome, Brave, Playwright or Puppeteer install, and its background daemon needs neither Playwright nor Node.js to operate.

A click fails early and names the specific element covering its target, such as a consent banner or modal, instead of silently clicking through it. An experimental WebMCP mode announces a page's available agent-callable tools automatically, with full schemas fetched only on request rather than sent proactively.

Rust, Apache-2.0 licensed.

More in Browser automation

See all 88
  • Loads an unpacked Chrome extension that lets a local AI coding agent click, type and read the tabs, cookies and logins of a Chrome window you already have open.

  • Hronaut

    hronaut.dev

    A visible, persistent Electron browser that exposes durable, named agent workspaces over MCP, so a coding agent's browser session stays open and inspectable independently of any one AI conversation.

  • Prompt OS

    github.com

    A desktop AI agent for Windows that executes terminal commands, manages files, writes and runs code, browses the web and reads the screen, acting on natural-language requests rather than only answering them.