How does agent-browser identify elements on a page?

It uses the 'snapshot -i' command to generate unique, short references (like @e1 or @e2) for interactive elements, which the AI can then use to click, fill, or select without needing complex CSS selectors.

How does it handle iframes?

Iframes are automatically inlined in snapshots, allowing the AI to see and interact with elements inside them as if they were part of the main page, without requiring manual context switching.

Does it support single-page applications (SPAs)?

Yes, because it uses a real Chrome/Chromium instance via CDP and includes specialized wait commands like 'networkidle' to ensure dynamic content is fully loaded before interacting.

Can I use this for websites that require login?

Yes, it supports multiple authentication methods including persistent profiles, an encrypted auth vault for credentials, and the ability to import session state directly from your local browser.

Agent Browser Automation

Name: Agent Browser Automation
Author: newsbreakDuadua9

bynewsbreakDuadua9

0•

Automatización de Navegadores

Automates web interactions including navigation, form filling, and data extraction using a Chrome-based CLI.

This skill empowers Claude to perform complex browser-based tasks directly through a specialized CLI interface. It enables the AI to navigate websites, interact with UI elements like buttons and forms via interactive snapshots, take screenshots, and handle advanced authentication flows including OAuth and persistent sessions. Whether you're automating repetitive web tasks, testing web applications, or scraping dynamic content, this skill provides the necessary tools to interface with the web programmatically and securely using Chrome DevTools Protocol (CDP).

Características Principales

01Full-page screenshots, PDF generation, and annotated captures

02Interactive element discovery with ref-based navigation (@e1, @e2)

03Command chaining and batch execution for high-performance workflows

040 GitHub stars

05Advanced authentication management via encrypted vaults and session persistence

06Automatic iframe handling and content inlining for seamless interaction

Casos de Uso

01Automated web application testing and bug reproduction

02Streamlining administrative tasks like form submissions or report generation

03Dynamic data scraping from sites requiring login or complex interaction

Características Principales

01Full-page screenshots, PDF generation, and annotated captures

02Interactive element discovery with ref-based navigation (@e1, @e2)

03Command chaining and batch execution for high-performance workflows

040 GitHub stars

05Advanced authentication management via encrypted vaults and session persistence