Skip to content

browser-use.mdx

Latest commit

 

History

History
107 lines (77 loc) · 3.59 KB

browser-use.mdx

File metadata and controls

107 lines (77 loc) · 3.59 KB
title Browser Use

Browser Use is the AI browser agent that empowers anyone to automate repetitive online tasks, no code required. By integrating with Kernel, you can run Browser Use Agents and automations with cloud-hosted browsers.

Adding Kernel to existing Browser Use implementations

If you already have a Browser Use implementation, you can easily switch to using Kernel's cloud browsers by updating your Browser definition.

1. Install the Kernel SDK

uv add kernel

2. Initialize Kernel and create a browser

Import the libraries and create a cloud browser session:

from kernel import Kernel
from browser_use import Browser, Agent

# Initialize Kernel client
client = Kernel(api_key="your-api-key")

# Create a Kernel browser session
kernel_browser = client.browsers.create()

3. Update your Browser definition

Replace your existing Browser initialization to use Kernel's CDP URL and display settings:

# Update your Browser definition to use Kernel's CDP URL
browser = Browser(
    cdp_url=kernel_browser.cdp_ws_url,
    headless=False,
    window_size={'width': 1024, 'height': 768},
    viewport={'width': 1024, 'height': 768},
    device_scale_factor=1.0
)
Browser Use supports a wide range of browser configuration parameters. See the full list in the Browser Use docs. When running on Kernel, remember that browsers must use one of Kernel’s supported viewport sizes and refresh rates—see Viewports for the supported configurations.

4. Create and run your agent

Use your existing Agent setup with the Kernel-powered browser:

# Use with your existing Agent setup
agent = Agent(
    task="Your automation task",
    llm=your_llm_instance,
    browser_session=browser
)

# Run your automation
result = agent.run()

# Clean up
client.browsers.delete_by_id(kernel_browser.session_id)
If you're using Browser Use versions `< 0.7.9`, you may need to use a [custom resize class](https://github.com/onkernel/create-kernel-app/blob/main/templates/python/browser-use/session.py) to correct viewport sizing for the browser session. Here's how to use it: 
agent = Agent(
    task="Your automation task",
    llm=your_llm_instance,
    browser_session=BrowserSessionCustomResize(cdp_url=kernel_browser.cdp_ws_url)
)

Quick setup with our Browser Use example app

Alternatively, you can use our Kernel app template that includes a pre-configured Browser Use integration:

npx @onkernel/create-kernel-app my-browser-use-app

Choose Python as the programming language and then select Browser Use as the template.

Then follow the Quickstart guide to deploy and run your Browser Use automation on Kernel's infrastructure.

Benefits of using Kernel with Browser Use

  • No local browser management: Run automations without installing or maintaining browsers locally
  • Scalability: Launch multiple browser sessions in parallel
  • Stealth mode: Built-in anti-detection features for web scraping
  • Session persistence: Maintain browser state across automation runs
  • Live view: Debug your automations with real-time browser viewing

Next steps