Browser Automation

Automate web browsing tasks from the command line using natural-language instructions. Supports navigating pages, clicking UI elements, extracting structured data, and taking screenshots using either a local Chrome instance or a remote Browserbase environment when API keys are configured.

SafeAutomationpeytoncasperv1.0.1

View on ClawHub

Use Cases

Navigate to a site and perform scripted UI actions (click buttons, fill forms) via natural language
Extract page data (titles, tables, fields) optionally using a JSON schema for structure
Take screenshots for verification, QA, or documentation of web flows
Discover page elements/selectors using an observe/query step before acting
Run the same automation locally during development and remotely in production using Browserbase

browser-automationclistagehandweb-scrapingscreenshots

Feature	Local	Browserbase
Speed	Faster	Slightly slower
Setup	Chrome required	API key required
Stealth mode	No	Yes
Proxy/CAPTCHA	No	Yes
Best for	Development	Production/scraping

Browser Automation

Use Cases

Browser Automation

First: Environment Selection (Local vs Remote)

Setup (First Time Only)

Commands

Quick Example

Mode Comparison

Best Practices

Troubleshooting