# Scrapling MCP Server [Health: Active]

**Category:** 🔎 Search & Data Extraction  
**Repository:** https://github.com/D4Vinci/Scrapling  
**GitHub Stars:** 100420  
**Views:** 9  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/scrapling-mcp-server

## Description
Web scraping with stealth HTTP, real browsers, and Cloudflare bypass. CSS selectors supported.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `uvx` (confidence: high):

```json
"mcpServers": {
  "scrapling-mcp-server": {
    "command": "uvx",
    "args": ["scrapling"]
  }
}
```

## Documentation

## What Scrapling MCP Server does

Scrapling MCP Server brings Scrapling's Python web-scraping framework into an MCP workflow for AI agents. It allows an agent to browse pages and extract information from websites rather than limiting scraping to manually written Python programs or terminal commands.

The underlying framework covers both simple page retrieval and larger crawling jobs. It can fetch static HTML with HTTP requests, use stealth browsers for JavaScript-heavy sites, interact with pages, and capture data from API calls made in the background. The server is suited to one-page extraction as well as workflows that need to traverse and collect content across a site.

## How it works

An agent uses the MCP interface to request web content and extraction operations through Scrapling. Retrieved pages can be queried with CSS selectors, XPath, text searches, or regular expressions. Scrapling also provides adaptive element tracking and matching for similar elements, which can help identify repeated structures in a page. Extracted content can be returned as clean Markdown when that format is more useful than raw page markup.

For broader collection tasks, Scrapling provides concurrent spiders with reusable sessions, proxy rotation, automatic retries, AutoThrottle, pause and resume support, and streaming. These capabilities make the same framework applicable to both targeted lookups and multi-page crawls.

## Tools and capabilities

- Fetch pages through fast HTTP requests or stealth browsers.
- Render JavaScript and interact with websites.
- Handle Cloudflare Turnstile and interstitial challenges.
- Capture API calls made in the background by pages.
- Select content with CSS, XPath, text, and regular expressions.
- Track adaptive elements and find similar page elements.
- Produce clean Markdown from scraped content.
- Run concurrent crawlers with proxy rotation, retries, throttling, and pause/resume support.
- Export crawl results as JSON, JSONL, CSV, or XML.

## Limitations and notes

Scrapling is intended for web scraping and extraction, so the quality of an agent's result depends on the target site's structure, accessibility, and behavior. Static pages can use ordinary HTTP fetching, while JavaScript-driven or interactive pages may require a browser-based fetcher. Proxy rotation, crawling controls, and export features are part of Scrapling's wider framework and are relevant when an MCP task expands beyond a single page.

The project can also be used directly from Python, through its command-line interface, or with an interactive shell. The MCP server is one access path alongside those interfaces, allowing an AI agent to use the same scraping framework.

_Full upstream README: https://allmcps.com/mcp/scrapling-mcp-server/readme_