devtools

URL Metadata / OG Preview

Preview the Open Graph card Slack, Discord, and social sites build from a URL. Our server fetches the page and reads its og:title, og:image, and meta tags.

How to use URL Metadata / OG Preview

What it does & when you need it

Paste a URL and this tool shows you the link preview that Slack, Discord, iMessage, WhatsApp, LinkedIn, and the social platforms will build when someone shares that page. That card — the big image, the bold headline, the one-line description — does not come from the visible page content. It comes from a small set of Open Graph <meta property="og:..."> tags in the page <head>: og:title, og:description, and og:image. Crawlers read those tags and ignore your actual body copy, which is why a page can look great in a browser and still produce an ugly or blank preview when shared.

Reach for it when you are shipping a marketing page, a blog post, or a launch announcement and want to confirm the card looks right before it hits a thousand timelines — or when a shared link already looks broken and you need to see what the crawler is actually reading.

Unlike most tools on this site, this one is not purely in-browser. To read a page's tags the way a real crawler does, the request has to come from a server, not from your browser (cross-origin rules would block a direct fetch anyway). So when you press Preview, the URL you enter is sent to our server, which fetches the page — capped at 512KB — parses the <head>, and returns the tags. If that privacy trade-off matters for an internal or unpublished URL, keep it in mind: the page is retrieved by us, exactly as a social scraper would retrieve it.

How to use

  1. Enter the full URL (including https://) in the Page URL field.
  2. Press Preview, or use Ctrl/Cmd + Enter. A loading line appears while we fetch the page.
  3. The top card is a rendered social preview: og:image, then the site name, the Open Graph title (falling back to the <title>), and the description.
  4. Below it, a raw table lists every field we extracted — title, description, og:title, og:description, og:image, and favicon — so you can spot exactly which tag is missing or wrong. Fields the page never set are shown as not set.

Things worth knowing

A missing og:image is the usual culprit behind an ugly card. If the tag is absent, or the image file is too large or too slow for the crawler to fetch in time, platforms fall back to the page <title> and a generic globe or document icon. That is why some links render as a sad grey box even though the page itself is fine — the crawler gave up on the image and moved on.

Use an absolute URL for og:image, ideally 1200×630. Relative paths like /og.png frequently fail because a scraper resolves them against a different base than a browser does, so it ends up requesting the wrong address and getting nothing. Always give the full https://… URL, and size the image at 1200×630 (a 1.91:1 ratio) so it fills the card on the widest surfaces without cropping.

Decoding is not verifying, and reading tags is not fetching assets. This tool reports the tag values; it does not download and re-encode the image or guarantee the crawler can reach it. If og:image looks correct here but the card is still blank, the image host may be blocking bots or serving it too slowly — check the response with the HTTP Header Viewer, and confirm the hostname resolves with the DNS Lookup.

Escaping matters inside meta tags. An ampersand or angle bracket in a title has to be written as an HTML entity, or the tag can break parsing; the HTML Entity Encoder is handy for cleaning those up before you paste them into a template. What you see in this preview is what sharers will see — the same page, fetched the same way a social crawler fetches it.

Examples

Preview a well-tagged site

https://github.com

GitHub sets a full set of Open Graph tags, so this returns a complete reference card with image, title, and description.

Inspect a content page's tags

https://developer.mozilla.org

See how a documentation site fills og:title and og:description, and check whether an og:image is present.

Frequently asked questions

Where does the link preview actually come from?

From the page's Open Graph meta tags — og:title, og:description, and og:image in the HTML head — not from the visible body content. Crawlers on Slack, Discord, iMessage, and the social platforms read those tags and ignore the rest, which is why the card can differ completely from what a reader sees.

Why is my link preview broken even though og:image is set?

The most common causes are an image that is too large or too slow for the crawler to fetch in time, or a relative path like /og.png. When the image fails, platforms fall back to the page title and a generic icon. Use an absolute https URL and keep the file reasonably small.

What size should the og:image be?

Aim for an absolute URL pointing to a 1200×630 image (a 1.91:1 ratio). That fills the large card on the widest surfaces without cropping, and an absolute URL avoids the resolution mismatch that makes relative paths fail for crawlers.

Does this tool send my URL to a server?

Yes. Reading a page's tags the way a real crawler does requires a server-side fetch, so the URL you enter is sent to our server, which retrieves the page (capped at 512KB) and parses the head. This is one of the few tools here that is not purely in-browser.

Why doesn't the preview match my page's headline?

Platforms prefer og:title over the HTML title tag and only fall back to title when og:title is missing. If your card shows the wrong headline, set an explicit og:title rather than relying on the visible page heading.