Workflow steps connected in a row, ending in a captured web page
Integrationsn8nAutomation

Automate website screenshots with n8n

A screenshot is rarely the end of a job. It goes to a folder, a channel, a report or a model. n8n is where those steps already live, and the Website Screenshot World node puts the capture in the same place.

· 8 min read

The idea

Why put screenshots in a workflow

Most screenshot work is glue. Take the picture, then do something with it: save it beside the invoice, post it when the pricing page changes, hand the page to an AI agent to read. Written as code, each of those is a small service to host and keep alive. In n8n it is a few nodes on a canvas.

The node covers both directions. It makes captures, as a step in any workflow, and it starts workflows when a capture finishes or a monitored page changes.

Setup

Installing the node

The node is verified by n8n, so on n8n Cloud it installs from the canvas. Open the nodes panel, search for Website Screenshot World, select it under More from the community and choose Install. An owner or admin of the instance does this once; every member can then use the node.

On self-hosted n8n the same search works, or open Settings → Community Nodes → Install and enter the package name:

@websitescreenshotworld/n8n-nodes-websitescreenshotworld

The n8n integration page has the steps and the full list of operations.

Setup

Three credentials, one per job

A workflow only holds the secret it uses.

  • Instant Snap API, for capture operations. A key from Developer Tools → API Keys in the app, starting with wsw_api_.
  • Auto Schedule API, for capturing a scheduled page on demand. A key from the same page, starting with wsw_autoapi_.
  • Webhook API, for the trigger. The signing secret from Developer Tools → Webhooks, starting with whsec_.

Test on a key checks it with the API. For the signing secret it only checks the shape, because a secret is never sent anywhere: it is used to check each delivery.

Capabilities

What the node can do

  • Take Screenshot. PNG, JPEG, WebP or TIFF, of the viewport or the full page.
  • Create PDF. An image PDF, the page as it looks on screen, or a document PDF with selectable, searchable text.
  • Record Video. MP4, WebM, MOV or GIF of the page scrolling from top to bottom.
  • Get Page Content. The main content of the page as Markdown or plain text, without navigation, footers or cookie banners.
  • Get. An earlier capture by its ID, with its file if you want it.
  • Capture Now. Captures a page that is scheduled in Auto Schedule right now, outside its schedule, with all its settings, comparisons and deliveries.

Output chooses between the file itself, as binary data ready for the next node, and the capture details with a download link. The options cover what the API offers: an emulated phone, viewport size, ad and pop-up blocking, a proxy country, an element to capture, elements to hide or click, request headers and cookies, and more. Anything left unset keeps the API's default, which blocks ads and pop-ups and captures a 1920 × 1080 desktop window.

Events

Starting a workflow from a capture

The trigger node starts a workflow when we send a webhook: a capture succeeded or failed, a scheduled capture differs from the one before it, an upload to cloud storage finished, a schedule was paused, or a sitemap run completed.

  1. Add the trigger to a workflow and choose the Webhook API credential.
  2. Copy the trigger's Production URL.
  3. In the app, open Developer Tools → Webhooks → Add endpoint, paste the URL and tick the events to send.
  4. For Auto Schedule events, switch on Trigger webhook under After the capture in the Web Profile the schedule uses.
  5. Activate the workflow.
The signature is checked for you
Every delivery is checked against its signature before the workflow starts. One with a missing, wrong or more than five minutes old signature is answered with 401 and starts nothing. If you want to see what that check involves, we wrote it up in how to verify a webhook signature.

Endpoints must be public https addresses, so an n8n running only on your own computer needs a tunnel to receive deliveries.

Examples

Four workflows to copy

  • Save a screenshot of every new article to Google Drive. RSS Feed Trigger → Take Screenshot with the item's link and Full Page on → Google Drive upload of the binary field data.
  • Tell Slack when a competitor's pricing page changes. Schedule the page in Auto Schedule with comparisons on, then: trigger on Capture Changed → If change_percent is above 5 → Slack.
  • Give an AI agent the web. AI Agent, with this node as a tool: Get Page Content to read pages as Markdown, and Take Screenshot to look at them.
  • Archive pages as PDFs. Webhook → Create PDF as an image PDF with Full Page on → Email.
Reliability

When the API refuses

A refusal stops the node with the API's own message and code, for example quota_exhausted, rate_limited or validation_failed. Turn on Retry On Fail in the node settings for the temporary ones, and set On Error to Continue to handle refusals in the workflow itself: the item then carries an error field. A delivery can also arrive more than once; its id stays the same, so use it to skip repeats where that matters.

Questions

FAQ

Is the node available on n8n Cloud?

Yes. n8n has verified it, which is what makes a community node appear in n8n Cloud's nodes panel: search for Website Screenshot World there and install it. It is also published on npm for any self-hosted n8n.

Does a screenshot taken in n8n cost extra?

No. It is an Instant Snap API capture made with your own key, so it counts against your plan and its rate limit like any other, and appears in your request log.

Can an n8n AI agent use it?

Yes. The node can be used as a tool by an n8n AI agent. Get Page Content gives the agent a page as Markdown or plain text, and Take Screenshot lets it look at the page.

How long is a captured file kept?

Files from the Instant Snap API are kept for 24 hours, and a download link works for 60 minutes. Choose the file output to pass the file straight to the next node, such as Google Drive, S3, email or Slack.

Why does my trigger not fire for scheduled captures?

An endpoint alone is not enough for Auto Schedule events. Switch on Trigger webhook under After the capture in the Web Profile the schedule uses.

Summary

TL;DR

Install the community node, add the credential for the job, and screenshots, PDFs, videos and page text become ordinary workflow steps. Add the trigger and a finished capture or a changed page starts a workflow of its own.