Playwright Scraper

Provides a Puppeteer-driven scraper that converts web pages to Markdown via a single tool, with optional SSL verification control.

python

0

GitHub Stars

python

Language

5 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "sudinigoutham-mcp_playwright_scraper": {
      "command": "uvx",
      "args": [
        "mcp-playwright-scraper"
      ]
    }
  }
}

This MCP server scrapes web pages and converts the resulting content into Markdown. It leverages a headless browser to handle modern, JavaScript-heavy sites, cleans up the HTML, and produces high-quality Markdown output you can store, share, or publish.

How to use

Configure the MCP client to run the server locally, then issue a scrape request by invoking the server’s tool to fetch a URL and return Markdown. The server exposes a single tool named scrape_to_markdown that requires the url parameter and optionally accepts a verify_ssl flag. Use it to capture the visible content of any web page and convert it to clean Markdown for documentation, notes, or content ingestion.

How to install

Prerequisites: install Python 3.11 or higher and ensure you can install Python packages.

Install using the Python package manager:

pip install mcp-playwright-scraper

Run the server in your environment using the module interface:

python -m mcp_playwright_scraper

Install Playwright browser dependencies so the scraper can render modern pages:

playwright install --with-deps chromium

If you plan to run the server with the uvx tool, ensure you have uvx available and then start the server using the appropriate command for your environment.

Additional content

Configuration and usage options are described for multiple integration environments. You can run the server via UVX or via Python, depending on your preferred workflow.

Example configurations for MCP clients are shown below to enable local execution of the server in different clients.

{
  "mcpServers": {
    "mcp-playwright-scraper": {
      "command": "uvx",
      "args": ["mcp-playwright-scraper"]
    }
  }
}

Troubleshooting and debugging

If you need to debug, use the MCP inspector with the appropriate start command for your setup to access the debugging URL and inspect requests and responses.

Available tools

scrape_to_markdown

Scrapes a URL using a headless browser, cleans the HTML, and converts the content to Markdown using Playwright, BeautifulSoup, and pypandoc.