Use cases

What people actually build with this

Situations that come up again and again, each written out in full: what goes wrong without a browser, what the API does about it, and the request that does the work.

Get your free API key Read the docs

AI and agents

For teams building agents

Web access for AI agents

Render the page an agent needs and hand it Markdown that reads like the page looks, from your code or as MCP tools.

Read the use case

For sites that want to be read by AI

llms.txt from your docs

List your documentation's pages, turn each into Markdown, and assemble the two files AI assistants look for.

Read the use case

For agents that need to see the page

Screenshots for vision models

Send a model the picture of the page together with each element's position, role, name and selectors, from one capture.

Read the use case

For tools that generate interfaces

Rendering generated HTML

Render the HTML a model generated at three screen sizes and hand the pictures back to it for review.

Read the use case

For developers working with coding agents

Coding agents checking their work

Give a coding agent a real browser so it can look at the preview of its own change, errors included, before it reports done.

Read the use case

For editors and curators

Newsletter curation

Read every article on your shortlist as clean Markdown, without the banners and pop-ups, ready to summarise.

Read the use case

For support and documentation teams

FAQ drafts from documentation

Turn a documentation page into question-and-answer pairs drawn from its own text, as validated JSON.

Read the use case

Data extraction and search

For teams doing this to a whole site

Crawling a whole site

Point a durable crawl at a seed URL, or hand a batch a list, and collect every page's content and a picture of it.

Read the use case

For teams feeding a database

Structured data from web pages

Name each field and its CSS selector, and get JSON back from the rendered page, with every field marked found or missing.

Read the use case

For pages with no reliable selectors

AI extraction to a JSON Schema

Send a page, a sentence about what you want and a JSON Schema, and get back JSON that has already been checked against it.

Read the use case

For teams adding search or answers

Semantic search over your content

Index pages or text with one call, then ask questions in plain language and get the closest passages back with their sources.

Read the use case

For pages that hide content until asked

Content behind clicks

Click the tab, accordion or load-more button first, then read the page, in one request.

Read the use case

For sales and market research

Company research from websites

Read a company's own pages and get its profile back as JSON in your schema, ready for the CRM.

Read the use case

For teams testing extraction and AI pipelines

Evaluation sets from real pages

Crawl a site you have the right to use into a fixed set of real pages, each with its Markdown and a screenshot to check labels against.

Read the use case

Monitoring and alerts

For teams watching something

Website change monitoring

Capture a page on a schedule, compare what it says rather than its markup, and get a signed webhook when it changes.

Read the use case

For readers and newsletters

A feed for a site without RSS

Read the site's index page on a schedule, keep the links you have seen, and turn the new ones into feed items.

Read the use case

For product and pricing teams

Competitor price tracking

Read a competitor's plan names and prices off the rendered page as values, and keep the history yourself.

Read the use case

For brands that sell through retailers

Minimum advertised price checks

Read each retailer's advertised price as a number and save a screenshot of the page beside it, as dated evidence.

Read the use case

For recruiters, researchers and job seekers

New job posting alerts

Watch a careers page and get a webhook when its list of job links changes.

Read the use case

For developers who depend on other people's APIs

Vendor changelog watching

Watch the changelogs and deprecation pages of the services you integrate with, and get a webhook when their text changes.

Read the use case

For teams that bid, apply or comply

Tender, grant and notice alerts

Watch notice and listing pages for new entries, and keep a full-page picture of the listing every time it is checked.

Read the use case

SEO and site health

For teams that ship websites

SEO regression checks

Read the head tags a search engine reads and fail the build when a deploy ships a noindex or a wrong canonical.

Read the use case

For teams whose links get shared

Link preview checks

Read the Open Graph and description tags a page publishes, served or rendered, and catch a broken card before it is shared.

Read the use case

For audits, migrations and sitemaps

Find every page on a site

Combine a site's sitemaps with the links on its pages, and see where each URL was found.

Read the use case

For teams moving a site

Website migration checks

Capture each old URL and its new counterpart together, and compare where it lands, what it says and what it declares.

Read the use case

For sites built on JavaScript frameworks

JavaScript SEO checks

Read a page as your server sends it and as a browser renders it, and compare what each one contains.

Read the use case

For teams with a performance budget

Page weight and load metrics

Get a page's resources with their sizes and lab load metrics from one capture, and compare them across deploys.

Read the use case

For teams that earn links

Backlink checks

Read the pages that should link to you and get back every link to your domain, and which of them are nofollow.

Read the use case

For sites that rely on rich results

JSON-LD structured data checks

Pull every JSON-LD block from the rendered page and check the fields your rich results depend on, on every deploy.

Read the use case

For sites in more than one language

hreflang checks

Read each language version's hreflang tags and confirm they all point at one another.

Read the use case

For agencies and freelancers

Website audits for prospects

Capture a prospect's home page at three screen sizes with its title, description and social tags, as the opening of a pitch.

Read the use case

Testing and QA

For teams that design for every screen

Responsive layout checks

One capture returns the page at phone, tablet and desktop size, ready to review side by side.

Read the use case

For teams that review UI changes

Visual regression testing

Capture a preview and production the same way on every build, with motion and moving parts removed, and compare the pixels yourself.

Read the use case

For teams that deploy front ends

JavaScript error checks

Load the page in a real browser and get back its console errors, uncaught exceptions and failed requests as JSON.

Read the use case

For flows that must keep working

Signup and checkout flow checks

Describe the steps a user takes and get a pass or fail for each one from a real browser, with a clear error when one breaks.

Read the use case

For designers and design-system teams

Hover, focus and dark mode screenshots

Hold a button in its hover or focus state, render the page in dark mode, and capture exactly that.

Read the use case

For teams shipping in more than one language

Localization screenshots

Capture each language at phone, tablet and desktop size, with the browser set to that language's locale and timezone.

Read the use case

For teams working on accessibility

Unlabelled control checks

List every control on the rendered page with the name a screen reader would announce, and find the ones that have none.

Read the use case

For walkthroughs, QA evidence and bug reports

A screenshot of every step

Run a flow of clicks and typing and get a screenshot before the first step and after every step that completes.

Read the use case

For products that run on customers' sites

Embedded widget checks

Load each customer's page, wait for your widget and capture only the widget, or learn that it never appeared.

Read the use case

For support and QA teams

Screenshots for support tickets

Capture the reported page at the customer's screen size, with its console errors and failed requests drawn on the image.

Read the use case

For teams running paid campaigns

Ad landing page checks

Load every ad destination on a phone in one batch, and see where it lands, what it answered and what it shows.

Read the use case

For sites rendered on the server and the client

Before and after hydration

Capture the page when its HTML first arrives and again once its scripts have run, side by side.

Read the use case

For QA engineers and automation builders

Selectors for test automation

Get every element on the rendered page with ranked selector candidates, from test IDs to roles and names.

Read the use case

For pages people print

Print layout checks

Render the page with its print styles applied, as an image or a PDF, and review it without a printer.

Read the use case

Screenshots, images and PDFs

For products full of links

Thumbnails and link previews

Turn every URL in a directory, bookmark list or search result into an image, without running a browser fleet.

Read the use case

For products that send documents

PDFs from your own HTML

Send HTML in the request and get a PDF back, laid out by real Chromium at the paper size you choose.

Read the use case

For products that draw their own images

HTML to image

Lay the image out in HTML and CSS, send it in the request, and get back a PNG at exactly the size you set.

Read the use case

For tools that save pages for later

Web page to PDF

Point the request at a URL and get a paginated PDF back, rendered after the page's own scripts have run.

Read the use case

For lists too long to loop over

Screenshots of a list of URLs

Submit up to 100 URLs in one request and collect each page's screenshot from a durable job.

Read the use case

For teams that document a product

Product screenshots for docs

Regenerate every screenshot in your help centre from the live app on each release, cropped to the part the article is about.

Read the use case

For teams that report from dashboards

Dashboard screenshots to chat

Capture a signed-in dashboard as an image on your own schedule and post it where your team already reads.

Read the use case

For agencies and freelancers

Monthly reports for clients

Capture each client's key pages as named, saved PDFs on a schedule, for the monthly report and the record.

Read the use case

For products that send or design email

Email template previews

Render each email's HTML to an image at email width, for template galleries, approvals and archives.

Read the use case

For products built around reviewing websites

Full-page captures for review tools

Capture the whole page, lazy-loaded images included, as one image your users can pin comments to.

Read the use case

Security, evidence and compliance

For teams that need a record

Web archiving and evidence

Save an image or PDF of what the browser saw, with the text alongside, as a side effect of work you already do.

Read the use case

For teams answering for their trackers

Third-party script audit

Load a page as a brand-new visitor and get the list of every outside domain it contacted before anyone clicked anything.

Read the use case

For security and trust teams

Suspicious link checks

Open a reported link in a remote browser and get a picture, the text, the redirect chain and every domain it contacted.

Read the use case

For teams that run public websites

Hacked site detection

Watch your key pages from outside and get a webhook when their links, text or title change with no deploy to explain it.

Read the use case

For journalists and researchers

Tracking silent edits

Watch a published page, keep the picture and the text of every version, and get told when the wording changes.

Read the use case

For marketing teams in regulated industries

Marketing page records

Capture each campaign landing page in full when it goes live, saved under a name your compliance team can find.

Read the use case

02 / 02 · Start here

Three ways in

Whichever of these is yours, it starts the same way.

Run one against a real page

The playground makes a live request and shows you the Markdown, the nodes, the image and the exact code to reproduce it. You can try it without an account.

Open

Read the guide

Quickstart, the request shape, field extraction, crawling, polling, and the endpoint reference - with the OpenAPI contract linked from every page.

Open

Wire it into your agent

The MCP server exposes capture and extraction as tools, so an assistant can open a page itself instead of being handed a scrape.

Open

PlansMarkdown extraction and image capture are on every plan, including the free one. Semantic snapshots and the rest of the DOM intelligence start at Pro; crawl and batch start at Business. The pricing page lists what each tier includes.

Give your product a browser.

Get clean web content and visual proof into your workflow in minutes.

Get your free API key Read the docs