Can I use Surf for mobile web testing?

Yes, Surf includes built-in emulation for various devices like iPhone and Pixel, as well as touch interaction and viewport resizing capabilities.

How does Surf find elements on a page?

Surf generates a simplified accessibility tree with unique references (refs) for interactive elements, but it also supports semantic locating by ARIA roles, labels, and text content.

Can Surf handle multi-window or incognito workflows?

Yes, Surf can manage multiple windows and tabs simultaneously, and it supports creating isolated incognito windows for clean-state testing.

Does Surf require API keys for AI services like ChatGPT or Gemini?

No, Surf interacts with AI services through your existing logged-in Chrome session, allowing you to use these models without needing individual API keys.

What happens if an AI service UI changes?

Surf includes validation tools like 'surf grok --validate' to check UI compatibility and can save updated model configurations to adapt to service changes.

Surf Browser Automation

Name: Surf Browser Automation
Author: w-winter

byw-winter

•

Automatización de Navegadores

Controls Chrome browsers via CLI to automate testing, web scraping, and AI-powered page analysis.

Surf is a comprehensive browser automation tool that allows Claude to interact directly with Chrome through a command-line interface or Unix socket. It provides deep inspection capabilities via accessibility trees, allowing for precise interaction with web elements using semantic refs, roles, or text. Beyond standard automation like clicking, typing, and scrolling, Surf uniquely integrates with web-based AI platforms such as ChatGPT, Gemini, and Grok by utilizing your existing browser sessions. This makes it an ideal solution for developers who need to perform complex E2E testing, capture automated screenshots, emulate various devices and network conditions, or leverage AI reasoning on live web content without requiring separate API keys.

Características Principales

01Native integration with browser-based AI assistants like ChatGPT, Gemini, and Perplexity.

02Robust multi-window and tab management including isolated incognito sessions.

03Advanced page inspection using accessibility trees and semantic element discovery.

0461 GitHub stars

05Complete browser control including navigation, clicking, typing, and scrolling via CLI.

06Comprehensive device, network, and CPU emulation for performance and mobile testing.

Casos de Uso

01Extracting and summarizing data from complex web applications using logged-in AI sessions.

02Automating end-to-end web testing and visual regression audits via screenshots.

03Debugging mobile-first layouts and network-throttled performance scenarios.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add w-winter/dot314 surf

For use in Claude.ai and ChatGPT

Download Skill