/tools/og-image-checker · free, no sign-up
OG image and link preview checker
Paste a URL and see the card X, LinkedIn, Slack and Discord build from it, with what to fix: the og:image size and format, twitter:card, title and description lengths, favicon, robots.txt for AI crawlers and llms.txt.
We fetch the page once, with its share image, favicon, robots.txt and llms.txt. We don’t save the report.
The tags a good share card needs
| tag | what to put in it |
|---|---|
| og:title | The product name and what it does, under about 60 characters. |
| og:description | One or two sentences, under about 160 characters. |
| og:image | An absolute https:// URL to a 1200×630 PNG or JPEG under 5 MB. |
| og:image:alt | What the image shows, for screen readers. |
| og:site_name | Your product or site name; Slack and Discord show it above the title. |
| twitter:card | summary_large_image, so X shows the image full width. |
<meta property="og:title" content="Acme — invoices that chase themselves">
<meta property="og:description" content="Acme sends your invoices and follows up until they're paid.">
<meta property="og:image" content="https://acme.com/og.png">
<meta property="og:image:alt" content="Acme's invoice list with three paid invoices">
<meta property="og:site_name" content="Acme">
<meta name="twitter:card" content="summary_large_image">
Questions
What size should an og:image be?
1200×630 pixels (a 1.91:1 ratio), as a PNG or JPEG under 5 MB. X, LinkedIn, Facebook, Slack and Discord all show that shape as a large card; smaller images are shown soft or as a small thumbnail, and other shapes are cropped. SVG is not shown.
Why does X show a small card or no image?
X needs <meta name="twitter:card" content="summary_large_image"> for a full-width image; "summary" gives a small square thumbnail. X reads og:title, og:description and og:image when the twitter: tags are missing. X retired its own Card Validator, so a checker like this one is the way to preview the card before you post.
I changed my og:image. Why do previews still show the old one?
Platforms cache previews. Give the new image a new URL (a version in the file name or query string), then use LinkedIn’s Post Inspector or Facebook’s Sharing Debugger to refresh their copy; Slack and Discord pick it up after their cache expires.
What is llms.txt?
A Markdown file at /llms.txt, proposed at llmstxt.org, that tells AI tools what a site is and which pages matter: an H1 with the name, a one-line “>” summary, then sections of links. It’s optional and no assistant is required to read it; what decides whether assistants can read your site at all is robots.txt.
Which AI crawlers should robots.txt allow?
To be found and cited, allow the search crawlers: OAI-SearchBot and ChatGPT-User (OpenAI), Claude-SearchBot and Claude-User (Anthropic), PerplexityBot, and Googlebot. GPTBot, ClaudeBot and Google-Extended collect training data; blocking them is a separate choice and doesn’t stop assistants citing you. A firewall or CDN bot block can still stop crawlers that robots.txt allows.
Do you store the pages I check?
No. We fetch the page, its share image, favicon, robots.txt and llms.txt once and show you the report; we don’t save it.
Link ready? Launch it.
Launchdog is a weekly product launch board, a Product Hunt alternative where you launch from your AI coding agent over MCP and people vote. Your agent drafts the listing from the same tags this page just read.