browser-use

Provision cloud browser sessions and run autonomous browser automation tasks.

Updated Aug 23, 2026

Search Tags:#automation #cloud #cdp #sessions #browser #profiles #web-scraping

Ready to run

npx skills add https://github.com/MagicWifiMoney/openclaw-starter-kit --skill browser-use-magicwifimoney

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: browser-use
Source: https://github.com/MagicWifiMoney/openclaw-starter-kit/tree/main/skills-custom/browser-use
Command: npx skills add https://github.com/MagicWifiMoney/openclaw-starter-kit --skill browser-use-magicwifimoney

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Provides a reliable, API-driven way to provision cloud browser sessions and run autonomous browser automation so you can perform login-protected workflows, web scraping, and interactive tasks without managing local browsers or complex tooling.

Core Features & Use Cases

Cloud Browser Sessions: Create and manage time-limited browser sessions that expose a CDP endpoint for direct control.
Persisted Profiles: Use profiles to retain cookies and login state across sessions for authenticated workflows.
Autonomous Tasks (Subagents): Submit prompts to dedicated browser-use LLMs to run scripted browsing tasks such as price checks, form automation, or data extraction.
Use Case: Start a profile-backed session, connect your agent to the session CDP URL, and run a task to scrape product listings behind a login.

Quick Start

Create a profile, start a browser session with that profile, and run an autonomous task to retrieve the price of a product from a target site.

Frequently Asked Questions about browser-use

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

FAQPage Schema

How do I run browser automation tasks in the cloud without managing local infrastructure?▼

Cloud browser automation provisions time-limited remote sessions exposing a CDP endpoint for direct control, enabling interactive tasks without local browsers. You create a session, connect via the CDP URL, and execute automated workflows.

Can I retain login states and cookies for authenticated web scraping across multiple sessions?▼

Persisted profiles retain cookies and login state across sessions for authenticated web scraping. By starting a browser session with a saved profile, your agent bypasses login walls to access protected content reliably.

How do I use LLM-driven subagents for autonomous web scraping and form automation?▼

Submit prompts to dedicated browser-use LLMs to run scripted browsing tasks like price checks, form automation, and data extraction. These autonomous subagents operate within your provisioned cloud browser session automatically.

What are the limitations of using time-limited cloud browser sessions for automated workflows?▼

Time-limited cloud browser sessions enforce billed usage constraints and require active polling for task completion. You must manage session lifecycles and ensure autonomous tasks complete before the provisioned session expires.

What's the best way to scrape product listings behind a login wall remotely?▼

Start a profile-backed browser session, connect your agent to the CDP URL, and submit an autonomous task prompt to extract product listings. This approach leverages retained cookies to bypass authentication for scraping.