n8n-nodes-hyper-reader

🦉 Agent-ready web scraper optimized for Claude, GPT-4, and Gemini

This is an n8n community node for Hyper-Reader - a high-fidelity, LLM-optimized web content extraction tool.

Features

• 🎯 Agent-Optimized Output - Pre-formatted for Claude, GPT-4, Gemini, or SearchGPT
• 🧹 85% Noise Reduction - Strips ads, navbars, tracking, and junk
• 🕵️ Elite Stealth Mode - Bypasses anti-bot protection
• 📸 Vision Support - Capture screenshots for Vision AI analysis
• 🔗 Deep Read - Follow internal links for broader context
• ⚡ Fast - 1-second response times via Standby Mode

Installation

Community Nodes (Recommended)

1. Go to Settings > Community Nodes
2. Click Install
3. Enter n8n-nodes-hyper-reader
4. Click Install

Manual Installation

cd ~/.n8n/nodes
npm install n8n-nodes-hyper-reader

Prerequisites

You need an Apify API token:

1. Create an account at apify.com
2. Go to Settings > Integrations
3. Copy your API token

Usage

Operations

Scrape URL

Extract clean content from a single URL.

{
  "url": "https://example.com/article",
  "title": "Article Title",
  "content": "# Article Title\n\nClean markdown content...",
  "metadata": {
    "author": "John Doe",
    "publishDate": "2026-01-15"
  },
  "wordCount": 1250,
  "tokensSaved": 8500,
  "processingTime": 1240
}

Scrape Multiple

Process multiple URLs in one run.

Deep Read

Scrape a URL and automatically follow internal links to gather more context. Perfect for:

• Documentation sites
• Multi-page articles
• Product catalogs

Get Run Status

Check progress of an async scraping job.

Agent Presets

| Preset | Description | |--------|-------------| | Claude | XML-structured Markdown with clear sections | | GPT-4 | Citation-heavy format with source references | | Gemini | Compact, efficient Markdown | | SearchGPT | Web-search optimized with key facts | | Raw | Clean Markdown, no agent optimization |

Stealth Levels

| Level | Use Case | |-------|----------| | 1 - Basic | Fast, for simple sites | | 2 - Standard | Good for most sites | | 3 - Elite | LinkedIn, Amazon, protected sites |

Workflow Examples

AI Research Assistant

[Webhook] → [Hyper-Reader] → [OpenAI] → [Google Sheets] → [Slack]

Scrape article → Summarize with AI → Log to vault → Notify team

Competitor Monitoring

[Schedule] → [Hyper-Reader (Multiple)] → [Compare] → [Alert]

Daily scrape → Compare changes → Alert on updates

RAG Pipeline

[Trigger] → [Hyper-Reader (Deep Read)] → [Embeddings] → [Vector Store]

Deep read docs → Generate embeddings → Store for retrieval

Vision Analysis

[Hyper-Reader (Vision)] → [GPT-4 Vision] → [Report]

Capture screenshot → Analyze with Vision AI → Generate report

Configuration Options

| Option | Default | Description | |--------|---------|-------------| | Agent Preset | Claude | Target AI optimization | | Output Format | markdown | markdown, json, or html_cleaned | | Stealth Level | 2 | Anti-bot protection (1-3) | | Capture Screenshot | false | Enable Vision mode | | Include Metadata | true | Title, author, date | | Preserve Links | true | Keep hyperlinks | | Include Images | false | Image URLs in output | | Exclude Selectors | nav, footer... | CSS selectors to remove | | Max Content Length | 0 | Truncation limit (0 = none) |

Deep Read Options

| Option | Default | Description | |--------|---------|-------------| | Depth | 1 | Link levels to follow (1-3) | | Max Pages | 10 | Maximum pages to extract |

Output Example

Claude Preset Output

<document>
<metadata>
<title>How to Build AI Agents</title>
<source>https://example.com/ai-agents</source>
<author>Jane Smith</author>
<date>2026-01-15</date>
</metadata>

<content>
# How to Build AI Agents

AI agents are autonomous systems that...

## Key Components

1. **Perception** - How agents understand their environment
2. **Reasoning** - Decision-making processes
3. **Action** - Executing tasks in the world

...
</content>
</document>

GPT-4 Preset Output

# How to Build AI Agents

Source: https://example.com/ai-agents
Author: Jane Smith | Published: January 15, 2026

AI agents are autonomous systems that... [1]

## Key Components

1. **Perception** - How agents understand their environment [2]
...

---
[1] Definition adapted from Russell & Norvig
[2] See also: Embodied AI research

Token Savings

Hyper-Reader strips ~85% of web page noise:

| Metric | Raw HTML | Hyper-Reader | Savings | |--------|----------|--------------|---------| | Characters | 150,000 | 22,500 | 85% | | Tokens (GPT-4) | ~37,500 | ~5,625 | ~$0.032/page |

Pricing

| Tier | Price | Best For | |------|-------|----------| | Standard | $1 / 1,000 pages | Blogs, news, docs | | Elite | $5 / 1,000 pages | LinkedIn, Amazon, protected sites | | Pro Monthly | $49 / month | Standby Mode, unlimited proxy |

Resources

• Hyper-Reader on Apify
• Apify Documentation
• n8n Documentation

Support

• Issues: GitHub Issues
• Author: Jason Pellerin (@ai_solutionist)
• Website: jasonpellerinfreelance.com

License

MIT License - see LICENSE for details.

Built with 🦉 by Jason Pellerin

Transform web chaos into agent-ready intelligence.