GhostTypes/cloudscraper-mcp-server

CloudScraper MCP Server

A Model Context Protocol (MCP) server that enables AI Agents to scrape information from various pages

Features

• Cloudflare Bypass: Automatically handles Cloudflare protection, utilizing cloudscraper
• Multiple Transport Protocols: Supports both stdio and HTTP transport for different use cases
• Clean Content Extraction: Returns structured, LLM-friendly content
• Docker Ready: Easy deployment with included Docker configuration
• Flexible Response Options: Choose between detailed responses or raw content only

Available Tools

scrape_url

Scrapes a URL and returns comprehensive response data including headers and metadata.

Parameters:

• url (string): The URL to scrape
• method (string, optional): HTTP method to use (default: "GET")

Returns:

• status_code (integer): HTTP response status code
• headers (object): Response headers
• content (string): Extracted page content
• response_time (number): Request duration in seconds

scrape_url_raw

Scrapes a URL and returns only the raw page content for simplified processing.

Parameters:

• url (string): The URL to scrape
• method (string, optional): HTTP method to use (default: "GET")

Returns:

• content (string): Raw page content

Installation

1. Clone the repository:

git clone https://github.com/yourusername/cloudscraper-mcp-server.git
cd cloudscraper-mcp-server

2. Install dependencies using uv:

uv sync

Configuration

The server supports two transport protocols depending on your use case:

Stdio Transport

Use stdio transport for direct integration with AI tools like Claude Code and VSCode:

uv run server.py

This is the default mode and works best for:

• Claude Code integration
• VSCode with MCP extensions
• Direct AI assistant communication
• Command-line usage

Claude Code Configuration

Add the server to Claude Code using the CLI:

claude mcp add cloudscraper-mcp \
  --type stdio \
  --command "uv" \
  --args "run" "server.py" \
  --directory "/path/to/cloudscraper-mcp-server"

VSCode/IDE Configuration

Add this configuration to your MCP settings file:

{
  "mcpServers": {
    "cloudscraper-mcp": {
      "type": "stdio",
      "command": "uv",
      "args": [
        "run",
        "server.py",
      ]
      "cwd":  "/path/to/cloudscraper-mcp-server/server.py"
    }
  }
}

HTTP Transport

Use HTTP transport for web-based integrations and automation platforms:

MCP_TRANSPORT=http
MCP_HOST=0.0.0.0
MCP_PORT=8000
uv run server.py

This mode is ideal for:

• n8n workflow automation
• Web-based AI applications
• API integrations
• Remote access scenarios

Environment Variables

Docker Deployment

For containerized deployment, see for complete Docker setup instructions including building, running, and using Docker Compose.