web-reading

Extract main article content from web pages using subagents and fallbacks.

5|2|Updated Feb 26, 2026

Search Tags:#web search #content extraction #subagent #web content #browser integration

Ready to run

npx skills add https://github.com/AxGord/claude-workflow --skill web-reading-axgord

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: web-reading
Source: https://github.com/AxGord/claude-workflow/tree/main/templates/skills/web-reading
Command: npx skills add https://github.com/AxGord/claude-workflow --skill web-reading-axgord

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill addresses the challenge of incorporating web content into conversations without overwhelming the main context with raw data.

Core Features & Use Cases

Web Content Extraction: Extracts the main article content from web pages with precision, ignoring navigation and ads.
Subagent-Based Fetching: Utilizes subagents for background processing, ensuring the main context remains clean.
Fallback Mechanisms: Provides fallbacks like Jina Reader for HTML pages and WebSearch for inaccessible content.
Browser MCP Integration: Offers Browser MCP as a last resort for JavaScript-rendered content.

Quick Start

Use the web-reading skill to fetch the main content of the article from 'https://example.com/article'.

Frequently Asked Questions about web-reading

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

FAQPage Schema

How do I extract the main article content from a web page without getting navigation and ads?▼

Web content extraction uses subagents to fetch pages and isolate the main article content, ignoring surrounding navigation and ads. This process ensures your main context remains clean by processing raw data in the background.

What is the best way to fetch JavaScript-rendered web content?▼

For JavaScript-rendered web content, the Browser MCP integration acts as a last resort fallback. It processes dynamic page elements after standard extraction and Jina Reader fallback mechanisms fail to retrieve the rendered article.

Why does web content extraction fail on some HTML pages and how is it handled?▼

Web content extraction fails on inaccessible HTML pages when standard fetching encounters restrictions. The skill handles this by falling back to Jina Reader, and subsequently using WebSearch, to retrieve the necessary article data.

Can I use subagents for web content fetching to keep the main context clean?▼

Yes, subagent-based fetching processes web content in the background, ensuring the main context remains clean. This mechanism isolates raw HTML data from your primary conversation while delivering the precise extracted article text.

Does web-reading work with Browser MCP for dynamic page rendering?▼

Yes, web-reading integrates with Browser MCP as a last resort to handle JavaScript-rendered content. This integration allows the extraction of dynamic web pages that standard HTML fetching cannot process.