
Better Fetch
A Model Context Protocol server that intelligently fetches and processes web content, transforming websites and documentation into clean, structured markdown with nested URL crawling capabilities.
README
Better Fetch - Advanced Web Content MCP Server
A powerful Model Context Protocol (MCP) server that intelligently fetches and processes web content with nested URL crawling capabilities. Transform any documentation site or web resource into clean, structured markdown files perfect for AI consumption and analysis.
🚀 Key Features
🕸️ Smart Web Crawling
- Nested URL Fetching: Automatically discovers and crawls linked pages up to configurable depth
- Single Page Mode: Option for simple single-page content extraction
- Domain Filtering: Stay within the same domain or allow cross-domain crawling
- Pattern Matching: Include/exclude URLs based on regex patterns
🧠 Intelligent Content Processing
- Content Cleaning: Removes ads, navigation, scripts, and other noise automatically
- Smart Section Detection: Identifies main content areas (
<main>
,<article>
,.content
) - Automatic Titles: Generates meaningful section headers based on page titles and URL structure
- Table of Contents: Creates organized TOC with proper nesting
📝 Advanced Markdown Generation
- Clean Formatting: Converts HTML to well-structured markdown
- Code Block Preservation: Maintains formatting for code snippets and technical content
- Link Preservation: Keeps all important links with proper markdown syntax
- Metadata Integration: Includes source URLs, generation timestamps, and site information
⚙️ Highly Configurable
- Crawl Depth Control: Set maximum levels to crawl (default: 2)
- Page Limits: Control maximum pages to process (default: 50)
- Timeout Settings: Configurable request timeouts
- Respectful Crawling: Built-in delays between requests
- Error Handling: Graceful handling of failed requests and invalid URLs
📋 Available Tools
1. fetch_website_nested
Comprehensive web crawling with nested URL processing.
Parameters:
url
(required): Starting URL to crawlmaxDepth
(optional, default: 2): Maximum crawl depthmaxPages
(optional, default: 50): Maximum pages to processsameDomainOnly
(optional, default: true): Restrict to same domainexcludePatterns
(optional): Array of regex patterns to excludeincludePatterns
(optional): Array of regex patterns to includetimeout
(optional, default: 10000): Request timeout in milliseconds
2. fetch_website_single
Simple single-page content extraction.
Parameters:
url
(required): URL to fetchtimeout
(optional, default: 10000): Request timeout in milliseconds
💡 Use Cases
📚 Documentation Processing
- API Documentation: Convert REST API docs, SDK guides, and technical references
- Framework Docs: Process React, Vue, Angular, or any framework documentation
- Library Guides: Extract comprehensive guides from library documentation sites
- Tutorial Series: Gather multi-part tutorials into single organized documents
🔍 Content Analysis & Research
- Competitive Analysis: Gather competitor documentation and feature descriptions
- Market Research: Extract product information from multiple related pages
- Academic Research: Collect and organize web-based research materials
- Knowledge Base Creation: Transform scattered web content into structured knowledge bases
🤖 AI Training & Context
- LLM Context Preparation: Create clean, structured content for AI model training
- RAG System Input: Generate high-quality documents for Retrieval-Augmented Generation
- Chatbot Knowledge: Build comprehensive knowledge bases for customer service bots
- Content Summarization: Prepare web content for automated summarization tasks
🛠️ Installation & Setup
Installing via Smithery
To install Better Fetch for Claude Desktop automatically via Smithery:
npx -y @smithery/cli install @flutterninja9/better-fetch --client claude
Prerequisites
- Node.js 18+
- npm or yarn
- MCP-compatible client (Claude Desktop, VS Code with MCP extension, etc.)
Step 1: Clone and Install
git clone https://github.com/yourusername/better-fetch.git
cd better-fetch
npm install
Step 2: Build the Project
npm run build
Step 3: Test the Server (Optional)
# Quick test
npm run dev
# Or run comprehensive tests
node test-mcp.js
Step 4: Configure Your MCP Client
For Claude Desktop:
Add to your claude_desktop_config.json
:
{
"mcpServers": {
"better-fetch": {
"command": "node",
"args": ["/absolute/path/to/better-fetch/dist/server.js"],
"env": {
"NODE_ENV": "production"
}
}
}
}
For VS Code MCP Extension:
{
"better-fetch": {
"command": "node",
"args": ["/Users/yourusername/better-fetch/dist/server.js"]
}
}
For Custom MCP Client:
{
"name": "better-fetch",
"command": "node",
"args": ["/path/to/better-fetch/dist/server.js"],
"stdio": true
}
📖 Usage Examples
Basic Documentation Crawling
Fetch all the web contents from this Flutter Shadcn UI documentation site:
https://flutter-shadcn-ui.mariuti.com/
Use nested fetching with a maximum depth of 3 levels and process up to 100 pages.
Advanced Configuration
Fetch content from the React documentation but exclude any URLs containing 'api' or 'reference' and only process pages containing 'tutorial' or 'guide':
URL: https://react.dev
Max Depth: 2
Exclude Patterns: ["/api/", "/reference/"]
Include Patterns: ["/tutorial/", "/guide/"]
Max Pages: 30
Single Page Extraction
Extract the content from this specific page only:
https://nextjs.org/docs/getting-started/installation
Use single page mode to avoid crawling related links.
📄 Sample Output
The server generates comprehensive markdown files with the following structure:
# Site Name Documentation
*Scraped from: https://example.com*
*Generated on: 2024-01-15T10:30:00.000Z*
## Table of Contents
- [Getting Started](#getting-started)
- [Installation](#installation)
- [Quick Start](#quick-start)
- [API Reference](#api-reference)
- [Core Functions](#core-functions)
---
## Getting Started
*Source: [https://example.com/getting-started](https://example.com/getting-started)*
[Clean markdown content here...]
---
## Installation
*Source: [https://example.com/installation](https://example.com/installation)*
[Installation instructions in markdown...]
For a complete example, refer to output.md
which demonstrates the server's output when processing a real documentation site.
🔧 Development
Project Structure
better-fetch/
├── src/
│ └── server.ts # Main server implementation
├── dist/ # Compiled JavaScript
├── test-mcp.js # Testing utilities
├── output.md # Sample output file
├── package.json
├── tsconfig.json
└── README.md
Available Scripts
npm run dev # Run in development mode with hot reload
npm run build # Compile TypeScript to JavaScript
npm run start # Run the compiled server
npm run clean # Clean dist directory
npm test # Run test suite
Testing Your Changes
# Interactive testing
node interactive-test.js
# Automated test suite
node test-mcp.js
# Manual JSON-RPC testing
echo '{"jsonrpc":"2.0","id":1,"method":"tools/list"}' | node dist/index.js
🚦 Performance & Limits
Default Limits
- Max Depth: 2 levels (configurable)
- Max Pages: 50 pages (configurable)
- Request Timeout: 10 seconds (configurable)
- Crawl Delay: 500ms between requests (respectful crawling)
Performance Tips
- Set appropriate
maxPages
limits for large sites - Use
includePatterns
to focus on relevant content - Enable
sameDomainOnly
to avoid external link crawling - Adjust
timeout
based on target site response times
🤝 Contributing
We welcome contributions! Please see our Contributing Guide for details.
Development Setup
- Fork the repository
- Create a feature branch:
git checkout -b feature/amazing-feature
- Make your changes and add tests
- Commit your changes:
git commit -m 'Add amazing feature'
- Push to the branch:
git push origin feature/amazing-feature
- Open a Pull Request
📜 License
This project is licensed under the MIT License - see the LICENSE file for details.
🆘 Support & Issues
- Bug Reports: GitHub Issues
- Feature Requests: GitHub Discussions
- Documentation: Check the Wiki
🙏 Acknowledgments
- Built with the Model Context Protocol SDK
- Powered by Cheerio for HTML parsing
- Markdown conversion by Turndown
Made with ❤️ for the AI and developer community
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.