# Scrapling MCP Server

> Scrapling MCP Server is an open-source MCP server for web search and scraping. MCP server for AI clients to scrape sites with HTTP and stealth browser sessions. It runs locally on your machine and is started by your AI client.

- **Kind:** MCP server
- **Pricing:** Open source
- **Platforms:** CLI, self-hosted, Linux, macOS, Windows
- **Website:** https://scrapling.readthedocs.io/en/latest/ai/mcp-server.html
- **Repository:** https://github.com/d4vinci/scrapling
- **GitHub stars:** 103,681
- **License:** BSD-3-Clause
- **MCP Registry name:** io.github.D4Vinci/Scrapling
- **Categories:** web search and scraping, browser automation
- **Listing updated:** 2026-11-14

## About

Scrapling MCP Server brings the Scrapling scraping library to AI chat clients and agents. It exposes seventeen tools for fetching pages, either as one-shot calls or inside persistent sessions.

One-shot tools fetch a single URL and clean up after themselves. Session tools keep cookies and settings, let the AI continue on the same tab, inspect network history, take screenshots, and run chained browser actions. Login state can be saved to and restored from a file.

It targets developers who want an assistant to collect data from dynamic or protected sites through conversation.

## Features

- **HTTP requests:** Fast requests with browser fingerprint impersonation.
- **Stealth browser fetch:** Renders JavaScript pages with fingerprint spoofing and Cloudflare challenge handling.
- **Persistent sessions:** Browser and request sessions that keep cookies across calls.
- **Saved login state:** Save and reload cookies, local storage and IndexedDB.
- **Page extraction:** Read pages as snapshot, HTML, Markdown or text.
- **Chained actions:** Combine clicks, typing and waits in one call.

## Good for

- Scrape dynamic JavaScript sites
- Reuse a logged-in session for extraction
- Capture screenshots of page elements
- Inspect network requests of a page
- Automate form filling during scraping

## Install

### Claude Code (Terminal)

```sh
claude mcp add scrapling -- uvx scrapling
```

### Claude Desktop (claude_desktop_config.json)

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

### Cursor (~/.cursor/mcp.json)

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

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "scrapling": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "scrapling"
      ]
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

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

### Other clients (your client's MCP settings (most use this shape))

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

## Tools

- `make_request`: One-shot HTTP request with fingerprint impersonation.
- `browser_fetch_once`: Render and scrape one URL with a stealth browser.
- `browser_open`: Open a persistent stealth browser session.
- `open_request_session`: Open a persistent HTTP request session.
- `close_session`: Close a browser or request session.
- `list_sessions`: List active sessions and their settings.
- `browser_save_state`: Save cookies and storage to a file.
- `browser_load_state`: Restore saved state into an open session.
- `browser_fetch`: Fetch a URL through an open browser session.
- `session_make_request`: Send an HTTP request through a request session.
- `browser_screenshot`: Capture the page or an element as an image.
- `browser_observe`: Return a short image sequence of page changes.
- `browser_extract`: Read current page content as snapshot, HTML, Markdown or text.
- `browser_actions`: Chain clicks, typing, waits and dialog replies.
- `browser_evaluate`: Evaluate script in the current page.
- `browser_network_requests`: List network requests made by the page.
- `browser_network_request`: Inspect a single network request.

## FAQ

### How many tools are there?

Seventeen, split into one-shot and session tools.

### How is it installed?

Registry packages are a PyPI package and a container image, both over stdio.

### What license applies?

BSD-3-Clause.

### Where are saved state files stored?

On the machine running the MCP server.

### Is Scrapling MCP Server free?

Yes. Scrapling MCP Server is open source, so you can use and self-host it for free.

### How do I add Scrapling MCP Server to Claude Code?

Run this command in your terminal: claude mcp add scrapling -- uvx scrapling

### What are the best alternatives to Scrapling MCP Server?

The closest alternatives to Scrapling MCP Server are Browser Use, Firecrawl MCP Server, SearXNG Search, Oya Browser and Mantis.

## Alternatives

- [Browser Use](https://ai-powered.com/mcp/browser-use): Browser Use is a freemium MCP server for browser automation.
- [Firecrawl MCP Server](https://ai-powered.com/mcp/firecrawl-mcp-server): Firecrawl MCP Server is a freemium MCP server for web search and scraping.
- [SearXNG Search](https://ai-powered.com/mcp/searxng-search): SearXNG Search is an open-source MCP server for web search and scraping.
- [Oya Browser](https://ai-powered.com/mcp/oya-browser): Oya Browser is a freemium MCP server for browser automation.
- [Mantis](https://ai-powered.com/mcp/mantis-2): Mantis is an open-source MCP server for browser automation.
- [NeuralVerge](https://ai-powered.com/mcp/neuralverge): NeuralVerge is an MCP server for web search and scraping.
- [AI-Powered directory](https://ai-powered.com/mcp/ai-powered-directory): AI-Powered directory is a free MCP server for developer tools.
- [Shopify MCP Server (any store)](https://ai-powered.com/mcp/shopify-mcp-server-any-store): Shopify MCP Server (any store) is a paid MCP server for e-commerce.

---
Source: https://ai-powered.com/mcp/scrapling-mcp-server
