RELATED 01
Browser Use MCP Server
Lets an assistant drive a browser with Puppeteer: navigate, click, fill forms, read pages, take screenshots.
40kGitHub stars
792/wknpm downloads
56popularity score
Scrapling MCP Server
Lets an assistant scrape pages over HTTP or a real browser, pick elements by CSS selector and take screenshots
By Karim shoair (individual) Python Local · stdio Browser automation
113 popularity score ★ 99k ↓ 920k/mo
What it does
Scrapling is a Python web scraping library by Karim Shoair (D4Vinci); its MCP server exposes the library's fetchers to an assistant. The assistant can make plain HTTP requests with browser fingerprint impersonation, load dynamic pages in Chromium, or use Scrapling's stealth browser, which its docs say can get past Cloudflare Turnstile and other anti-bot checks. Pages come back as Markdown, HTML or clean text, and a CSS selector can narrow the result to the elements needed before it reaches the model, which saves tokens.
There are thirteen tools: one-shot tools that each start and close their own client or browser (single and bulk versions), and session tools that keep a browser or HTTP session open across calls, plus a screenshot tool. Sessions can also connect to an already-running remote browser through a Chrome DevTools Protocol URL. Browser tools block about 3,500 ad and tracker domains, and responses are stripped of hidden content that could carry prompt injection.
It runs locally: install with pip install "scrapling[ai]" and run scrapling install for the browser dependencies, or use the Docker image pyd4vinci/scrapling (also on ghcr.io). It speaks stdio by default; a Streamable HTTP mode (--http) is available and requires an auth token since version 0.4.15. Proxies and a custom Chromium executable can be configured.
What it needs
| Runtime | Python |
|---|
Tools 13
make_request- HTTP request (GET, POST, PUT, DELETE) with browser fingerprint impersonation.
bulk_get- Async GET of several URLs at once.
fetch- Fetch dynamic content with a Chromium/Chrome browser.
bulk_fetch- Fetch several URLs in parallel browser tabs.
stealthy_fetch- Fetch with Scrapling's stealth browser to get past Cloudflare Turnstile/Interstitial and other anti-bot systems.
bulk_stealthy_fetch- Stealth fetch of several URLs in parallel tabs.
open_session- Open a persistent browser session (dynamic or stealthy), optionally on a remote browser via CDP.
open_request_session- Open a persistent HTTP session keeping cookies, connections and fingerprint.
close_session- Close a browser or HTTP session and free its resources.
list_sessions- List active sessions with their settings.
session_fetch- Fetch one URL through an open browser session.
session_make_request- Make an HTTP request through an open HTTP session.
screenshot- Capture a PNG or JPEG screenshot of a page in an open browser session.