web-browser

Control Chrome via DevTools Protocol to automate browsing tasks.

1|Updated Jan 27, 2026

Search Tags:#cdp #chrome #ai-agent #web-automation #remote-debugging #chromium

Ready to run

npx skills add https://github.com/devskale/skale-skills --skill web-browser-devskale

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: web-browser
Source: https://github.com/devskale/skale-skills/tree/main/skills/web-browser
Command: npx skills add https://github.com/devskale/skale-skills --skill web-browser-devskale

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This skill enables automated web browsing and interaction by remotely controlling Chrome/Chromium through the Chrome DevTools Protocol, removing the need for manual repetitive web tasks for AI agents.

Core Features & Use Cases

Click buttons, fill forms, and navigate links on web pages.
Retrieve page data (titles, content) and capture screenshots for further processing.
Use in AI agent workflows to explore, test, and collect data from websites.

Quick Start

Start Chrome with remote debugging enabled (port 9222) and a fresh profile using the provided start script:

node scripts/start.js
node scripts/start.js --profile Navigate to a URL with the navigation script:
node scripts/nav.js https://example.com Run arbitrary JavaScript in the active page:
node scripts/eval.js "document.title" Take a screenshot of the current page:
node scripts/screenshot.js

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 browsing and form filling with Chrome DevTools?▼

Automate web browsing by remotely controlling Chrome/Chromium through the Chrome DevTools Protocol. The skill connects via WebSocket to a Chrome instance with remote debugging enabled on port 9222, allowing you to click buttons, fill forms, navigate links, and extract page data programmatically.

What do I need to set up to start remote debugging with Chrome?▼

Start Chrome or Chromium with remote debugging enabled on port 9222 using the provided start script (node scripts/start.js). The skill requires a fresh Chrome profile and the ws Node.js library to establish a WebSocket connection for remote control.

Can I capture screenshots and retrieve page content during web automation?▼

Yes. The skill lets you take screenshots of the current page state and retrieve page data like titles and content. Run node scripts/screenshot.js to capture visuals or node scripts/eval.js to execute arbitrary JavaScript and extract information.

How do I navigate to URLs and run JavaScript commands remotely?▼

Use node scripts/nav.js [url] to navigate to a page and node scripts/eval.js [command] to execute JavaScript in the active page. Both scripts communicate with Chrome via the Chrome DevTools Protocol to perform the actions and return results.

What are the limitations when using remote Chrome debugging for web automation?▼

The skill requires Chrome/Chromium with remote debugging explicitly enabled on port 9222—it cannot work with standard browser instances. Large-scale automation across many pages may face latency or connection limits depending on your Chrome setup and network conditions.

Can I use this for AI agent workflows to explore and test websites?▼

Yes. The skill integrates into AI agent workflows for automated site exploration, testing, and data collection. It removes manual repetition by letting agents click, navigate, fill forms, and gather data across diverse web environments through programmatic Chrome control.