How does this skill handle dynamic web content?

The skill uses ARIA snapshots to generate an accessibility tree, allowing the AI to find elements by their role and name rather than brittle CSS selectors. It also includes specific wait commands for selectors and URL patterns.

How do I perform complex scraping without excessive scrolling?

For large datasets, the skill supports intercepting and replaying API requests. This allows you to capture the underlying data source and schema directly rather than parsing the DOM.

What is the purpose of the 'release' command?

The release command is critical because it unlocks the user interface and clears the 'Agent Operating' indicator in Max, allowing the human user to resume interaction with the browser page.

Can I run TypeScript code inside the browser?

No. Code passed to the browser's execution context must be plain JavaScript. TypeScript syntax will fail at runtime as the browser does not natively understand it.

Browser Automation

Name: Browser Automation
Author: maxgent-ai

bymaxgent-ai

0•

Browser Automation

Automates web browser interactions and data extraction with persistent session state for complex, multi-step workflows.

The Browser Automation skill enables Claude to perform sophisticated web-based tasks by maintaining a persistent page state across multiple command executions. Built on Playwright, it allows for precise navigation, form filling, and element interaction using an ARIA-based accessibility tree for reliable discovery. It is ideal for developers needing to automate website testing, scrape data via API interception, or perform repetitive browser workflows that require human-like logic and visual verification through screenshots.

Key Features

01ARIA-based element discovery for robust interaction with dynamic layouts

02Visual debugging tools including full-page screenshot capabilities

03Advanced web scraping via API request interception and replay

04Persistent browser state across sequential command executions

05Direct JavaScript execution within the browser context

060 GitHub stars

Use Cases

01Automating repetitive administrative tasks like form submissions and logins

02Automating end-to-end testing for web applications

03Extracting structured data from complex, paginated websites

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add maxgent-ai/maxgent-plugin browser

For use in Claude.ai and ChatGPT

Download Skill