Extract Metadata, Markdown and Screenshots from any URL

Get link metadata — title, description, images and favicon — for rich previews, clean Markdown for LLMs and screenshots for everything else. One API key for all of it.

Get free API key Read the docs

100 free credits every month · No credit card · 2,600+ developers signed up

Live demo

Try:

GET /api/extract · 1 credit

Five APIs, one key

Shared credits across every endpoint. Use what you need, when you need it.

Link Metadata API

Extract metadata from any URL: title, description, images, favicon and site name — ready for rich link previews.

1 credit · GET /api/extract →

URL to Markdown

The main content of any page as clean Markdown — no nav, ads or boilerplate. Built for RAG and LLM context.

1 credit · POST /api/markdown →

Screenshot API

Full-page or viewport captures in PNG or JPG, retina-ready, with cookie banners removed.

2 credits · GET /api/screenshot →

Link Extractor API

Every link on a page, with its anchor text — for crawlers, SEO audits and site maps.

1 credit · GET /api/links →

Image Extractor API

Every image on a page, including lazy-loaded ones — for galleries, scraping and previews.

1 credit · GET /api/images →

Coming soon

AI Extract

Describe the fields you need in plain English and get structured JSON back from any page.

POST /api/ai-extract

FOR LLM PIPELINES

Feed your model the content, not the markup

Raw HTML wastes tokens on scripts, menus and tracking. JsonLink pulls out the article and hands back Markdown with its title, language and word count.

Readable Markdown with headings, lists and links kept
Works on JavaScript-heavy pages through a real browser
Same key as metadata and screenshots

page.html

<div class="nav-wrap"><ul>…
<script src="gtm.js"></script>
<div id="cookie-modal">…
<article>
 <h1>How DNS works</h1>
 <p>Every lookup starts…
<aside class="ad-slot">…
<footer>…

markdown

# How DNS works

Every lookup starts with
a recursive resolver…

## Root servers

- 13 named roots
- [RFC 1034](https://…)

Simple credit pricing

Metadata and Markdown cost 1 credit. A screenshot costs 2. Credits renew every month.

Free

100 credits / month

Hobbie

$15/mo

50,000 credits / month

Growth

$49/mo

200,000 credits / month

Scale

$99/mo

500,000 credits / month

Compare plans →

Free online tools

No signup needed. Powered by the same API.

Extract images from URLFind and download every image on a page Link preview generatorSee how a link looks when shared Link extractorExport every link from a web page Metadata scraperCheck a page's title, description and OG tags

Frequently asked questions

What is JsonLink?

An API that takes any URL and returns its metadata, its main content as Markdown, or a screenshot — as JSON your app can use right away.

How do I get metadata from a URL?

Call /api/extract with the URL and your API key. You get title, description, images, favicon, site name and domain.

What metadata can I extract from a link?

JsonLink reads Open Graph, Twitter Card and standard HTML meta tags and returns the page title, description, preview images, favicon, site name and domain — the same link metadata used by Slack, WhatsApp or Twitter previews.

Can I convert a web page to Markdown for an LLM?

Yes. POST /api/markdown returns the readable content of the page as Markdown, with the title, language and word count.

Is there a free plan?

Yes. Every account gets 100 free credits every month, with no credit card. That's 100 metadata or Markdown requests, or 50 screenshots.

Start extracting in a minute

Get free API key