MCP For Claude Marketplace
Browserless (Playwright Cloud)

Browserless (Playwright Cloud) MCP Connector for Claude

A+

Equip your AI with a remote headless browser to scrape, interact, and run Playwright safely via cloud.

10 tools Official Updated Jun 28, 2026 Official Vinkius Partner
More Details Connect to Claude

Grant your AI agent the absolute power of an untethered, scalable Browserless / Playwright Cloud cluster. Unlike basic REST fetchers, this MCP connector spins up a legitimate remote Chromium/Webkit instance. Your agent can now execute full JavaScript lifecycles natively, scrape heavy SPAs (Single Page Applications), wait for CSS selectors, or simply ask for a high-fidelity PDF render of any site.

What you can do

  • Remote Scraping — Pull down the fully rendered HTML DOM of any complex site (get_html_content) bypassing typical bot-blocks. Intercept specific payload slices selectively using explicitly mapped CSS queries (scrape_elements).
  • Visual Operations — Command the agent to spawn an ephemeral browser and snap an immediate get_screenshot of a live staging site or construct a print-perfect get_pdf document autonomously.
  • Evasion & Stealth — Avoid standard Cloudflare/Data Dome constraints by wrapping requests in scrape_with_stealth to emulate human environments, or bounce TCP endpoints completely out utilizing a residential scrape_with_proxy.
  • Live Execution — Embed dynamic lifecycle alterations into the target memory structure. Run scrape_with_js to click modals inside the remote page prior to extraction, or run_custom_function exposing the raw Chrome DevTools Protocol to your LLM.

How it works

  1. Append the Browserless integration to your AI settings
  2. Provide your securely managed Browserless HTTP Token
  3. Start fetching full visual DOM objects via prompts inside Claude or Cursor

Who is this for?

  • QA Automation Teams — instruct the AI to spin up ephemeral browser endpoints rendering and screenshotting daily builds silently.
  • Data Engineers — execute data extractions dynamically reading through complex React/Vue frontend pages that actively combat standard scraping.
  • Security Researchers — proxy malicious URL queries entirely through isolated cloud environments avoiding all local machine contamination.
headless-browsertest-automationweb-scrapingchromiumbrowser-testingcloud-execution

10 tools expose this connector's capabilities to your AI agent.

get_html_content

Get raw HTML content using Playwright Cloud (Browserless)

send_custom_payload

Send custom JSON payload to any Browserless REST endpoint

run_custom_function

Execute custom JS function inside remote Playwright browser

scrape_with_js

evaluate` lifecycle executing imperative DOM manipulations like button clicks prior to scraping the final mutated HTML state. Inject and run a custom JS script before extracting content

get_pdf

Print page to PDF using Playwright Cloud

scrape_with_proxy

Scrape URL using a specified proxy server

scrape_elements

Extract specific elements from page using CSS selectors

get_screenshot

Take a full-page screenshot using Playwright Cloud

scrape_with_stealth

Scrape using stealth mode plugins to avoid bot detection

scrape_with_wait

Scrape content and wait for a CSS selector

See how to talk to your AI agent using Browserless (Playwright Cloud).

Take a full page screenshot of the landing page on 'https://vinkius.com'.

Screenshot execution completed. Spun up remote Playwright instance requesting `https://vinkius.com` in Full-Page context. It successfully rendered and extracted the composite `.png` buffer holding the pixel mapping. The remote artifact link has been provided [Graphic Representation File]. Is there another view you want audited?

Scrape the price of the first item from Amazon using a stealth bypass on `url_here`.

Stealth routine enacted via `scrape_with_stealth`. Bypassing WAF headers and standard bot challenges successfully on the retail endpoint. I subsequently applied `.a-price-whole` matching selectors via logical isolation. The payload returned values precisely: `["399", "419"]`. Validating extraction logic completed.

Retrieve the rendered HTML content of this site: 'https://vinkius.com/blog'.

I've invoked the `get_html_content` method successfully against the target URI. The Playwright container fully let the React/Next.js hydration loop stabilize. Captured massive 450kb parsed DOM structure explicitly mapping all internal blog routes natively. How would you like me to process this massive markup string within our memory context?

Absolutely. Because Browserless exposes actual Chrome environments via Playwright, standard JavaScript rendering pipelines execute exactly as they do on a user's machine. The agent simply captures the stabilized final DOM states post-hydration, making it remarkably effective for highly dynamic single-page applications.

Related Connectors

PI

PagerDuty Incident Trigger MCP

1 tools Official

This MCP does exactly one thing: it triggers incidents in PagerDuty. That's its only function, and nothing else. Incredible for giving your AI agents the power to wake up engineers.

F View details →
IN

InnoVint MCP

10 tools Official

Manage wine production, lots, vessels, lab analyses, and cellar actions for your InnoVint winery through natural conversation.

A+ View details →
GR

Granola MCP

12 tools Official

Manage AI meeting notes via Granola — list and search meeting documents, retrieve transcripts and summaries, and track action items directly from any AI agent.

A+ View details →
FU

FunnelCockpit MCP

10 tools Official

Build high-converting sales funnels, landing pages, and email sequences with an all-in-one German marketing platform.

A+ View details →