browser-use/terminal

Terminal UI to get stuff done in the browser

What it solves

It automates repetitive browser-based tasks by providing a terminal-based interface for AI agents to control a web browser, allowing users to execute complex workflows without manual navigation.

How it works

Built with Rust, the project uses a custom LLM harness and a Ratatui-based terminal UI to manage the agent loop and state. It controls browsers (real Chrome, headless Chromium, or Browser Use cloud) via the Chrome DevTools Protocol (CDP), giving the AI model direct access to page JavaScript, screenshots, and files. It also includes an SQLite event log to track history, artifacts, and traces.

Who it’s for

Users who want to automate browser tasks from their terminal, as well as developers using coding assistants (like Claude Code or Codex) that can execute shell commands to trigger browser actions.

Highlights

  • Flexible Browser Support: Works with logged-in Chrome sessions for account state, headless Chromium, or cloud-based browsers.
  • Coded for Speed: Uses a new LLM harness designed to be twice as fast and twice as cheap as Browser Harness.
  • Interactive Control: Allows users to watch, steer, stop, retry, and resume tasks in real-time via the TUI.
  • Assistant Integration: Can be plugged into coding assistants via a skill installation process, allowing them to execute browser scripts and view screenshots.

Related

  • Project
    keon/browser-control

    A fast Rust CLI that lets AI coding agents control a real browser via the Chrome DevTools Protocol using shell-native commands and stable element references.

  • Project
    vercel-labs/agent-browser

    A fast, native Rust CLI for browser automation tailored for AI agents, featuring accessibility tree snapshots and agent-friendly text extraction.

  • Project
    sleepinginsummer/agent-browser-cli

    A CLI tool that connects AI Agents to a real Chrome browser session, allowing them to reuse login states and perform precise page control and perception.

  • Project
    ultraworkers/claw-code

    A Rust-based CLI agent harness that allows users to run AI coding agents locally with support for multiple LLM providers.

  • Project
    lahfir/agent-desktop

    A native Rust-based desktop automation CLI that enables AI agents to interact with any application via OS accessibility trees, featuring token-efficient UI traversal and session tracing.