n8n has no built-in Puppeteer or Playwright node: n8n’s integrations documentation lists neither. To drive a real browser from a workflow, for pages that need JavaScript, screenshots, PDFs, logins or clicks, you install a community node: one that runs Puppeteer or Playwright next to n8n, or one that sends the work to a Browserless service. This guide compares them, using the description and README each author publishes on npm and our own download data.
What do you need?
| You want to | Look at |
|---|---|
| Screenshots, PDFs, page HTML or your own Puppeteer script | n8n-nodes-puppeteer |
| No browser on your n8n server, a hosted or self-hosted Browserless instead | n8n-nodes-browserless-api, or n8n-nodes-puppeteer with a WebSocket endpoint |
| Firefox or WebKit as well as Chromium, with click and form steps | n8n-nodes-playwright |
| Several Playwright steps on one remote browser session | n8n-nodes-playwright-core |
Headless browser community nodes
Our Puppeteer and Playwright hub lists 33 packages that pass our quality rule, with 24,753 npm downloads a week between them. The most downloaded, with the description each author gives on npm (weekly npm downloads as of 28 October 2026):
| Package | Weekly downloads | Latest version | Author’s description |
|---|---|---|---|
| n8n-nodes-puppeteer | 21,961 | 1.5.0 (23 January 2026) | Browser automation using Puppeteer |
| n8n-nodes-browserless-api | 1,493 | 1.2.1 (16 April 2026) | Community node for the Browserless REST APIs |
| n8n-nodes-browserless | 462 | 1.1.3 (3 October 2024) | Integration with the Browserless platform |
| n8n-nodes-playwright | 376 | 0.2.21 (2 January 2026) | Browser automation using Playwright |
| n8n-nodes-puppeteer-extended | 175 | 0.1.0 (25 October 2022) | Requesting webpages using Puppeteer |
| n8n-nodes-playwright-core | 60 | 2.0.0 (17 June 2026) | Browser automation with Playwright Core, browser agnostic |
Versions and dates come from the npm registry.
n8n-nodes-puppeteer
The most installed package by far. Its README lists four operations: get the full HTML of a page, capture a page as a PDF, take a screenshot of all or part of a page, and run a custom script with the full Puppeteer API plus the features of n8n’s Code node. Options include device emulation, extra headers, timeouts, a Wait Until setting, headless mode, a stealth mode, a human typing mode, launch arguments and capturing files downloaded during a script.
Puppeteer needs a browser. The README offers two setups:
- Local browser: a ready-made Docker setup in the repository with Chromium and its fonts and libraries, which the README recommends for production. In Docker or Kubernetes the node adds the Chrome launch arguments a container needs by itself. Without Docker, Puppeteer downloads its own Chromium; the README suggests installing system Chrome instead and setting
PUPPETEER_SKIP_CHROMIUM_DOWNLOAD=true. - Remote browser: set Browser WebSocket Endpoint to a managed service such as Browserless or to your own browser container, and pick the protocol: CDP for Chrome or Chromium, WebDriver BiDi for Firefox.
If you see errors about missing shared libraries such as libnss3.so, the README says to install the missing Chrome dependencies or switch to a remote browser. It also describes use as a tool for n8n’s AI Agent, where the agent can write the script with $fromAI().
n8n-nodes-browserless-api
Its repository is in the browserless GitHub organization. Nothing runs a browser on your n8n server: the node calls the Browserless REST APIs. Its README lists Get Content, Scrape (CSS selectors), Smart Scrape, Screenshot, PDF, Search, Map URLs, Run Function (your own JavaScript or Puppeteer code, run on the server), Download, Export, Unblock (CAPTCHAs and bot detection), Performance Audit (Lighthouse) and Crawl. The credential takes a Browserless API token and a base URL, which you change to your own address for a self-hosted Browserless.
n8n-nodes-browserless
An older Browserless node from another author, last released in October 2024. Its README says it supports the Browserless API V2 and lists rendering, JavaScript injection, PDF download, screenshots and custom code, and starts its setup with running Browserless in Docker.
n8n-nodes-playwright
Runs Playwright inside n8n with a choice of Chromium, Firefox or WebKit, headless mode and a slow motion option. Its README lists Navigate, Take Screenshot, Get Text, Click Element and Fill Form, with CSS or XPath selectors, and Run Custom Script with the full Playwright API. The package downloads its browser binaries at install, about 1 GB of disk space, and the README lists Node.js 18.10 or newer and n8n 1.0.0 or newer, and notes that extra system dependencies may be needed.
n8n-nodes-playwright-core
A fork that connects to a remote browser through a Playwright WebSocket endpoint, Chromium or Firefox only, and keeps a session open so several Playwright nodes in one execution can work on the same browser, for example to sign in on one step and extract data on the next. Its README lists Navigate, Get Text, Click Element, Fill Form, Take Screenshot, Download File, Run Custom Script and Close Session. Run Custom Script is off by default; an administrator turns it on with N8N_PLAYWRIGHT_NODE_CUSTOM_SCRIPT_ENABLED=true.
n8n-nodes-puppeteer-extended
Based on n8n-nodes-puppeteer, with clicks, form filling, CSS selector screenshots, HTML to JSON and sequences of actions in one session, per its README. Its only release dates from October 2022.
Check each node page here for the latest release and, where it has one, its GitHub repository for open issues before you build on a package.
How to choose
- Most workflows on a self-hosted n8n you control: n8n-nodes-puppeteer, the most used package, with its Docker setup or a remote browser.
- You can’t or don’t want to install a browser next to n8n: n8n-nodes-browserless-api, or n8n-nodes-puppeteer pointed at a Browserless WebSocket endpoint.
- Firefox or WebKit, or simple click and form steps without code: n8n-nodes-playwright.
- Logged-in flows across several nodes on a remote browser: n8n-nodes-playwright-core and its sessions.
- Scraping and crawling as a service instead of a browser: see Firecrawl in n8n and Best n8n Community Nodes for Apify.
To install a package on a self-hosted instance, an owner or admin enters its npm name in Settings > Community Nodes; full steps are in How to Install n8n Community Nodes. On n8n Cloud only verified community nodes can be installed, see Verified vs Unverified n8n Community Nodes. For environment variables such as the ones above, see n8n Community Node Environment Variables.
FAQ
Does n8n have a Puppeteer or Playwright node?
Not a built-in one. Both come as community nodes, the most used being n8n-nodes-puppeteer and n8n-nodes-playwright.
Why does the Puppeteer node fail with a missing library error?
The browser lacks system libraries such as libnss3.so. The n8n-nodes-puppeteer README says to install the missing Chrome dependencies, use its Docker setup, or connect to a remote browser with a WebSocket endpoint.
Can I use a headless browser in n8n without installing Chrome?
Yes, through a remote browser: n8n-nodes-browserless-api calls Browserless over its REST APIs, and n8n-nodes-puppeteer and n8n-nodes-playwright-core can connect to a browser over a WebSocket endpoint.
What is the most used headless browser community node?
n8n-nodes-puppeteer, with 21,961 weekly npm downloads on 28 October 2026.
