browser-use

Automates browser navigation, interaction, and data extraction via a persistent CLI daemon.

Updated Apr 1, 2026
One-click install
npx skills add https://github.com/fauzanazz/katalis --skill browser-use-fauzanazz
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: browser-use
Source: https://github.com/fauzanazz/katalis/tree/main/.agents/skills/browser-use
Command: npx skills add https://github.com/fauzanazz/katalis --skill browser-use-fauzanazz

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Manually driving a browser for testing, form filling, or scraping is slow and error-prone. This Skill gives an AI agent direct control of a real browser through the browser-use CLI, with a persistent daemon that keeps the browser open across commands for low-latency automation. ## Core Features & Use Cases - Full Browser Control: Navigate pages, click and type into elements by index, scroll, manage tabs, take screenshots, and wait for selectors or text. - Flexible Browser Modes: Run headless Chromium, connect to the user's real Chrome with existing logins, attach a specific Chrome profile, or provision a zero-config cloud browser. - Data Extraction & Sessions: Execute JavaScript, read element text and attributes, manage cookies, and run multiple isolated browser sessions in parallel. - Use Case: Ask the agent to log into a site using your Chrome profile, fill out a multi-field form, submit it, and capture a screenshot of the confirmation page. ## Quick Start Use the browser-use skill to open https://example.com, list the clickable elements on the page, and take a screenshot.

Frequently Asked Questions about browser-use

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate browser interactions from the command line?▼

Use the browser-use CLI: run browser-use open <url> to launch a headless browser, then browser-use state to list clickable elements with indices. Interact using commands like browser-use click 5 or browser-use input 3 "text", and verify with browser-use screenshot.

How do I use my existing Chrome login sessions for automation?▼

Run browser-use connect to attach to your running Chrome with remote debugging enabled, preserving cookies and logins. Alternatively, use browser-use --profile "Default" open <url> to launch a managed browser with your Chrome profile data.

Can I run multiple browser sessions in parallel?▼

Yes, pass --session NAME on every command to target an isolated browser instance with its own daemon, socket, and state. For example, run a cloud browser in one session and an authenticated Chrome session in another simultaneously.

Does browser-use support cloud browsers?▼

Yes, browser-use cloud connect provisions a zero-config cloud browser with a persistent profile and connects via CDP. It requires an API key set via browser-use cloud login or the BROWSER_USE_API_KEY environment variable.

Why does my browser-use command fail or target the wrong browser?▼

Failures usually come from a broken session; run browser-use close first, then retry. If commands hit the wrong browser, you likely forgot the --session flag, which defaults to the 'default' session.

How do I extract data from a web page with browser-use?▼

Use browser-use get text <index>, get html, or get attributes to read element content, or browser-use eval "js code" to run arbitrary JavaScript. For advanced needs like network interception, use browser-use python with raw CDP access.