web-access

Control a Chrome browser via CDP proxy for web research and page interaction.

Updated Apr 16, 2026

Search Tags:#authentication #web-scraping #browser-automation #remote-debugging #web-access #cdp-proxy

Ready to run

npx skills add https://github.com/jacardl/New-Radar --skill web-access-jacardl

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: web-access
Source: https://github.com/jacardl/New-Radar/tree/main/microservices/web-access
Command: npx skills add https://github.com/jacardl/New-Radar --skill web-access-jacardl

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill enables AI agents to perform web research, fetch content, and interact with websites through a local browser automation bridge. It provides end-to-end control of a Chrome-based session via a CDP proxy, allowing deterministic actions without leaving the AI environment.

Core Features & Use Cases

Local browser control via a CDP proxy to perform searches, fetch pages, and automate user interactions.
Support for login-protected sites and anti-scraping scenarios using a real-browser session for authenticated access.
Exposes HTTP endpoints (e.g., /new, /navigate, /eval, /screenshot) to drive deterministic web tasks; includes environment checks and guidance for setup.

Quick Start

Start the CDP Proxy, ensure Chrome remote debugging is enabled, then use the HTTP API to control pages.

Frequently Asked Questions about web-access

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

FAQPage Schema

How do I automate web scraping on login-protected sites using an AI agent?▼

Browser automation for AI agents works by exposing HTTP endpoints like /new, /navigate, /eval, and /screenshot through a CDP proxy. This setup allows deterministic control over a local Chrome session to fetch content and execute web research tasks.

What do I need to set up local browser control via a CDP proxy?▼

You need Node.js v22+ and a Chrome instance with remote debugging enabled to run the CDP proxy script. The proxy exposes HTTP endpoints that allow your AI agent to execute deterministic web page interactions and data retrieval.

Can I use browser automation to extract content and take screenshots programmatically?▼

Yes, you can extract content and take screenshots by sending requests to the /eval and /screenshot HTTP endpoints exposed by the CDP proxy. This enables your AI agent to capture page data and visuals directly from the local Chrome session.

Does web scraping with a real browser session help bypass anti-scraping mechanisms?▼

Yes, using a real Chrome browser session via a CDP proxy helps navigate anti-scraping scenarios. It enables authenticated access and executes web interactions through a genuine browser environment, making automated retrieval more reliable.

What are the limitations of using a CDP proxy for web page interaction?▼

Using a CDP proxy for web page interaction requires a local Chrome instance with remote debugging enabled and Node.js v22+. It is limited to deterministic tasks and depends on the active browser session's stability for consistent automation results.