skills.rest
wsxwj123
wsxwj123COMMUNITY
Explore all skills by this author →
GitHub

web-access

Manage authenticated browser-based web tasks via CDP proxy automation.

7|Updated Jan 27, 2026
Search Tags:#automation#chrome#web-search#data-extraction#browser-automation#web-access#cdp-proxy
One-click install
npx skills add https://github.com/wsxwj123/opencode-skills-backup --skill web-access-wsxwj123
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: web-access
Source: https://github.com/wsxwj123/opencode-skills-backup/tree/main/skills/web-access
Command: npx skills add https://github.com/wsxwj123/opencode-skills-backup --skill web-access-wsxwj123

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

AI agents struggle to access live information and interact with websites that require real browser sessions, authentication, and dynamic content. This skill provides a robust, policy-compliant solution that orchestrates browser-based tasks and web data retrieval for automation, testing, and research. Its design ensures safe, auditable interactions when performing online operations.

Core Features & Use Cases

  • CDP proxy-based browser automation for authenticated sessions and dynamic pages.
  • Supports multiple tooling modes: WebSearch, WebFetch, curl, and Jina for flexible data gathering.
  • Three interaction modalities: JS-based click, real mouse click via CDP, and file uploads.
  • Parallel sub-agent orchestration for multi-site investigations.
  • Site-patterns references to reuse domain-specific best practices.
  • Media extraction: locate and extract images/videos from DOM and render frames.

Quick Start

Install this skill and run a test CDP-enabled browsing task to fetch live data from a target site.

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 access for AI agents to retrieve live data from dynamic pages?▼

Automated web access for AI agents is achieved through a CDP proxy that manages real browser sessions, enabling authenticated, real-time data retrieval and dynamic page interaction. It supports WebSearch, WebFetch, curl, and Jina for flexible gathering.

Can I access login-protected content using browser automation with an AI agent?▼

Yes, login-protected content access is supported via CDP proxy-based browser automation. This approach manages authenticated sessions for AI agents, ensuring safe, auditable interactions while performing automated navigation and data extraction tasks.

What is the best way to orchestrate parallel web data extraction across multiple sites?▼

Parallel web data extraction across multiple sites is best handled through parallel sub-agent orchestration. This mechanism allows AI agents to conduct multi-site investigations simultaneously, applying site-pattern references to reuse domain-specific best practices.

Does this web automation approach support JS-based clicks, real mouse clicks, and file uploads?▼

Yes, the web automation approach supports three interaction modalities: JS-based clicks, real mouse clicks via CDP, and file uploads. These options enable flexible navigation and interaction with complex DOM elements during live information gathering.

How do I extract images and videos from the DOM during automated web navigation?▼

Media extraction from the DOM is performed by locating and extracting images and videos directly from the page structure. The automation skill can also render frames, allowing AI agents to capture visual media during live website testing and data gathering.

When should I not use CDP proxy for browser automation in web scraping tasks?▼

CDP proxy browser automation might not be suitable for simple, static data extraction where lightweight tools like WebFetch or curl suffice. It is designed for complex, authenticated, and dynamic web tasks requiring real browser sessions and auditable interactions.

Related Skills

Explore More Software Engineering Skills→
calvingitcalvingit
2

web-access

Coordinate search, fetch, login, and interactive browsing via a CDP proxy.

Community
Advanced
xberg-ioxberg-io
26

extracting-metadata

Extract titles, descriptions, language, and Open Graph tags from HTML documents.

Official
Intermediate
whichguywhichguy

concurrent-evolution

Coordinate eight concurrent sub-agent tests and generate a Sovereign Tournament Report.

Community
Advanced
phanijappsphanijapps
1

light-panda-browser

Automate web scraping and data extraction with the Lightpanda headless browser.

Community
Intermediate