web-browser

Automate Chrome/Chromium web interactions via the Chrome DevTools Protocol.

Updated Jul 10, 2022
One-click install
npx skills add https://github.com/samuerio/.dotfiles --skill web-browser-samuerio
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: web-browser
Source: https://github.com/samuerio/.dotfiles/tree/main/agents/skills/web-browser
Command: npx skills add https://github.com/samuerio/.dotfiles --skill web-browser-samuerio

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires ws, and includes scripts (resource) components.

What problem does it solve?

Interact with web pages by programmatically performing actions such as clicking, filling forms, and navigating links through remote control of Chrome/Chromium via the Chrome DevTools Protocol (CDP).

Core Features & Use Cases

  • Start Chrome with remote debugging enabled
  • Navigate to URLs, execute JavaScript, and take screenshots
  • Dismiss cookie dialogs across sites and collect logs for debugging
  • Interactive element selection and automated page exploration

Quick Start

Launch Chrome with the provided start script, then use nav.js, eval.js, pick.js, and screenshot.js to automate common web tasks.

Frequently Asked Questions about web-browser

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate web tasks using Chrome CDP?

Automate web tasks using Chrome CDP by remotely controlling Chrome/Chromium through the DevTools Protocol to click elements, fill forms, and navigate URLs programmatically. The Skill provides scripts like nav.js and eval.js to execute these interactions.

What's the best way to dismiss cookie banners during automated web testing?

Dismiss cookie banners during automated web testing by using the provided scripts to programmatically detect and interact with cookie dialogs across diverse sites. The Skill handles cookie-banner dismissal while collecting logs for debugging.

Do I need to launch Chrome with remote debugging enabled to extract web data?

Yes, you need to launch Chrome or Chromium with the --remote-debugging-port=9222 flag enabled to extract web data. The Skill interfaces via a minimal CDP client that requires this remote debugging port to function.

Can I take screenshots and execute JavaScript when automating web interactions?

Yes, you can take screenshots and execute JavaScript when automating web interactions. The Skill exposes screenshot.js for capturing page states and eval.js for executing custom JavaScript within the remote browser environment.

How does interactive element selection work for web page exploration?

Interactive element selection for web page exploration works through the pick.js script, which interfaces with the Chrome DevTools Protocol to identify and select elements on the page for automated interaction and testing.

What are the limitations of automating web interactions via a minimal CDP client?

The minimal CDP client relies on the ws dependency and requires a manually launched Chrome instance with remote debugging enabled. It does not manage the browser lifecycle automatically and is limited to interacting with exposed scripts under the scripts/ directory.