Fetcher MCP Server

Browser AutomationTypeScript

Fetch web page content using Playwright headless browser

Available Tools

fetch_url

Retrieves web page content from a specified URL using Playwright headless browser with intelligent content extraction

urltimeoutwaitUntilextractContentmaxLengthreturnHtmlwaitForNavigationnavigationTimeoutdisableMediadebug

fetch_urls

Batch retrieves web page content from multiple URLs in parallel using multi-tab fetching for improved performance

urlstimeoutwaitUntilextractContentmaxLengthreturnHtmlwaitForNavigationnavigationTimeoutdisableMediadebug

Fetcher is a powerful web content retrieval tool that uses Playwright's headless browser to fetch and process web pages. Unlike traditional web scrapers, it fully executes JavaScript, making it capable of handling dynamic web content and modern web applications. It features intelligent content extraction with a built-in Readability algorithm that automatically removes ads, navigation, and other non-essential elements from web pages. The tool offers flexible output formats (HTML or Markdown), parallel processing for batch operations, and configurable parameters for fine-tuned control. It optimizes resource usage by blocking unnecessary elements like images and stylesheets, and includes robust error handling to ensure reliable operation even with problematic web pages.

Installation

You can install and run Fetcher MCP in several ways:

Using npx (Quickest Method)

Run directly with npx:

npx -y fetcher-mcp

For first-time setup, install the required browser:

npx playwright install chromium

Docker Installation

Run with Docker:

docker run -p 3000:3000 ghcr.io/jae-jae/fetcher-mcp:latest

Or deploy with Docker Compose by creating a docker-compose.yml file:

version: "3.8"

services:
  fetcher-mcp:
    image: ghcr.io/jae-jae/fetcher-mcp:latest
    container_name: fetcher-mcp
    restart: unless-stopped
    ports:
      - "3000:3000"
    environment:
      - NODE_ENV=production
    volumes:
      - /tmp:/tmp
    healthcheck:
      test: ["CMD", "wget", "--spider", "-q", "http://localhost:3000"]
      interval: 30s
      timeout: 10s
      retries: 3

Then run:

docker-compose up -d

Configuration

Starting the Server

Start the server with HTTP and SSE transport:

npx -y fetcher-mcp --log --transport=http --host=0.0.0.0 --port=3000

This provides two endpoints:

/mcp - Streamable HTTP endpoint (modern MCP protocol)
/sse - SSE endpoint (legacy MCP protocol)

Debug Mode

To show the browser window for debugging:

npx -y fetcher-mcp --debug

Configuring in Claude Desktop

Add this configuration to your Claude Desktop config file:

On MacOS: ~/Library/Application Support/Claude/claude_desktop_config.json On Windows: %APPDATA%/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "fetcher": {
      "command": "npx",
      "args": ["-y", "fetcher-mcp"]
    }
  }
}

Usage Tips

Handling Special Websites

For websites with anti-crawler mechanisms, include in your prompt:

Please wait for the page to fully load

For slow-loading websites:

Please set the page loading timeout to 60 seconds

Content Retrieval Adjustments

To preserve original HTML structure:

Please preserve the original HTML content

To fetch complete page content:

Please fetch the complete webpage content instead of just the main content

To return content as HTML:

Please return the content in HTML format

Enabling Debug Mode

To display the browser window during a specific fetch operation:

Please enable debug mode for this fetch operation