Does it support mobile browser testing?

Yes, it features built-in support for iOS Simulators, enabling Claude to launch Mobile Safari, perform touch gestures like swipes, and take mobile-specific screenshots.

How does it manage dynamic content or slow-loading pages?

It provides advanced wait commands that can trigger on network idle states, specific URL patterns, or the appearance of particular elements to ensure the automation remains robust.

Can this skill handle authenticated sessions?

Yes, the skill includes state management commands that allow you to log in once, save the authentication state to a JSON file, and reload it for future automated sessions.

What is the benefit of the snapshot system in agent-browser?

The snapshot system generates unique reference IDs (like @e1) for interactive elements, allowing Claude to interact with specific UI components accurately without needing to maintain complex or brittle CSS selectors manually.

Agent Browser Automation

Name: Agent Browser Automation
Author: rodolfolermacontreras

byrodolfolermacontreras

•

ブラウザ自動化

Automates web interactions including navigation, form filling, data extraction, and stateful authentication for AI agents.

This skill empowers Claude to perform complex web-based tasks through a specialized browser automation CLI. It allows the agent to navigate websites, interact with interactive elements via reference IDs, capture snapshots, manage authenticated sessions, and even simulate mobile devices. Whether for automated testing, data scraping, or routine web workflows, this skill provides a structured framework for programmatic browser control with built-in state persistence and visual debugging capabilities.

主な機能

01Visual debugging tools including full-page screenshots and video recording

02Authentication state persistence to save and reuse login sessions

03Multi-session management for parallel browsing and data collection

04Mobile Safari simulation via iOS Simulator integration for mobile testing

05Interactive element referencing (@refs) for precise navigation and clicks

061 GitHub stars

ユースケース

01Automating complex multi-step web forms and authenticated site logins

02Extracting structured data and capturing visual snapshots for reporting

03Executing end-to-end web application tests directly from the CLI environment

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add rodolfolermacontreras/microsoft-ai-learning agent-browser

For use in Claude.ai and ChatGPT

主な機能

01Visual debugging tools including full-page screenshots and video recording

02Authentication state persistence to save and reuse login sessions

03Multi-session management for parallel browsing and data collection

04Mobile Safari simulation via iOS Simulator integration for mobile testing

05Interactive element referencing (@refs) for precise navigation and clicks

061 GitHub stars

ユースケース

01Automating complex multi-step web forms and authenticated site logins

02Extracting structured data and capturing visual snapshots for reporting

03Executing end-to-end web application tests directly from the CLI environment

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add rodolfolermacontreras/microsoft-ai-learning agent-browser

For use in Claude.ai and ChatGPT