Check Meta Tags: 5 Free Ways to Audit Any Page
November 11, 2026 · 11 min read
How to Check Meta Tags on Any Web Page: 5 Free Methods
To check meta tags on a page, look in three places: the raw HTML, the rendered page after JavaScript runs, and the HTTP response headers. View source, Chrome DevTools and curl cover those layers for free. Then confirm what Google, Facebook and LinkedIn actually fetched with their own inspection tools.
What to Look For When You Check Meta Tags
You can tell whether a website has meta tags by opening its source and searching for <meta. Nearly every page has some, so the useful question is whether the right ones are present, appear once, and say what you expect.
Use this checklist each time you check meta tags on a page. It tests presence and uniqueness, not wording.
| Tag | Where it lives | Pass if |
|---|---|---|
title | Head | Present once, unique to the page |
meta name="description" | Head | Present once, not copied from another page |
meta name="robots" | Head, or the X-Robots-Tag header | No noindex on a page you want in search |
link rel="canonical" | Head, or a Link header | One absolute URL that returns 200 |
meta name="viewport" | Head | Present on every page |
meta charset | Head | Present, near the top of the head |
html lang | The opening html tag | Matches the page language |
og:title, og:description, og:image, og:url | Head | All four present, absolute URLs |
twitter:card | Head | Present if you share links on X |
Skip the meta keywords tag. Google's special tags documentation says it isn't used by Google Search and has no effect on indexing or ranking. Length rules are a separate topic: Google truncates titles to fit the device width, not at a fixed character count. Our post on the HTML title tag covers how to check title length.
How Do I Find My Meta Tags with View Source?
View source shows the HTML exactly as your server sent it, before any script runs. In Chrome, press Ctrl+U on Windows or Linux, or Cmd+Option+U on a Mac. You can also type view-source: in front of the URL in the address bar.
Press Ctrl+F (Cmd+F on a Mac) and search for name=, property= and canonical. A healthy head looks something like this:
<html lang="en">
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>Pricing | Example Co</title>
<meta name="description" content="Plans, prices and limits for Example Co.">
<link rel="canonical" href="https://example.com/pricing">
<meta property="og:title" content="Example Co pricing">
</head>
Each line answers one checklist row: the language, the character set, mobile scaling, the search title, the snippet text, the preferred URL and the social title. Notice what's missing here: no robots tag, so no restrictions apply and search engines may index the page and follow its links.
This raw view matters most for link previews. Some social scrapers don't run JavaScript, so a tag that only exists after scripts run may never reach them.
Inspect the Rendered Head in Chrome DevTools
DevTools shows the live DOM, which includes anything JavaScript added after the page loaded. Open it with F12 or Ctrl+Shift+I on Windows and Linux, or Cmd+Option+I on a Mac. In the Elements panel, expand the head node to see every tag.
Scrolling through a long head gets tedious, so paste this into the Console tab instead:
const rows = [...document.querySelectorAll('meta')].map(m => ({
tag: m.getAttribute('name') || m.getAttribute('property') ||
m.getAttribute('http-equiv') || (m.hasAttribute('charset') ? 'charset' : ''),
value: m.getAttribute('content') || m.getAttribute('charset') || ''
}));
console.table(rows);
console.log('title:', document.title);
console.log('canonical:', document.querySelector('link[rel="canonical"]')?.href);
It prints a table with a tag and a value column for every meta tag on the page, then the title and the canonical URL. A blank tag cell usually means the element uses an attribute the snippet doesn't read, such as itemprop.
Now compare the two views. A tag that shows up in DevTools but not in view source was injected by JavaScript. Google can render JavaScript, but you shouldn't assume every crawler or link scraper does.
DevTools also exposes the third layer. Open the Network panel, reload, click the first document request, and read its Response Headers. Look for X-Robots-Tag and Link, because neither appears anywhere in the HTML.
How Can I Extract Meta Tags from a URL with curl?
curl fetches a URL from the command line and prints the raw HTML, which makes it the quickest way to pull tags from many pages. It ships with macOS, most Linux distributions, and, per the curl project, Windows 10 and 11. This pipeline prints only the title, meta and canonical tags:
curl -sL https://example.com/pricing \
| grep -ioE '<title[^>]*>[^<]*</title>|<meta [^>]*>|<link [^>]*rel="?canonical"?[^>]*>'
-s hides the progress bar and -L follows redirects. grep -o prints each tag on its own line, even in minified HTML. Many online meta tags checkers start with this same raw fetch. Extra title lines usually come from inline SVG icons, which carry their own title elements; the page title is the first match.
To audit several pages at once, put one URL per line in a file called urls.txt (your sitemap is a good source) and loop over it:
while read -r url; do
echo "== $url"
curl -sL "$url" | grep -ioE '<title[^>]*>[^<]*</title>|<meta [^>]*name="description"[^>]*>'
done < urls.txt
Each URL prints its title and description underneath, so a missing or repeated description stands out quickly. Then check the headers:
curl -sIL https://example.com/pricing | grep -iE '^(HTTP/|x-robots-tag:|link:)'
This shows the status code of every hop plus any robots or canonical header. A line like X-Robots-Tag: noindex means the page is blocked from search even if its HTML head looks clean. Google's robots meta tag documentation confirms that any rule you can put in the meta tag also works in this header.
Windows has no built-in grep. Type curl.exe (in Windows PowerShell 5.1, plain curl is an alias for Invoke-WebRequest) and pipe to findstr /i "x-robots-tag link:" for headers. Expect findstr to print whole lines, so minified HTML comes back as one long line.
The limitation: curl doesn't run JavaScript, and some sites block it or serve different HTML to different user agents. If its output disagrees with your browser, trust DevTools for the rendered result and curl for the raw one.
Test Meta Tags for Social Previews: Facebook Sharing Debugger and LinkedIn Post Inspector
Each social network fetches your page with its own crawler and caches the result. Testing meta tags in their own tools shows what each network actually stored, which can differ from what your page serves today. Most of them build the card from the tags the Open Graph protocol defines.
Meta's Sharing Debugger works as an og tags checker: you need a Facebook login, and it shows the tags Meta's crawler scraped plus any errors or warnings. Meta's webmaster docs say running a URL through it also triggers a fresh scrape, so it's how you update a preview after fixing your HTML.
LinkedIn's Post Inspector does the same for LinkedIn. Enter a URL and confirm the preview shows the right title, description and image. LinkedIn's help center says this refreshes the cached preview, but only for new posts. Posts that already include the link keep the old card.
X uses its own card tags and has its own testing quirks, so treat it as a separate check. Our post on the Twitter Card validator shows how to test X cards now.
Validate Meta Tags Google Sees with Search Console URL Inspection
On a site you own, URL Inspection lets you validate meta tags against what Google itself fetched. Paste a URL into the search bar at the top of Search Console. The indexed result shows the user-declared canonical (yours) and the Google-selected canonical (Google's choice).
Click Test live URL, then View tested page. Google's help page says this shows the HTML returned, a screenshot of the rendered page, the HTTP headers and JavaScript console output. The live test won't predict which canonical Google picks, so compare the two canonical fields on the indexed result.
For a page you don't own, Google's Rich Results Test accepts any public URL and lets you browse the rendered code it tested.
One caution: correct tags don't guarantee Google shows them. Google's title link documentation says it generates titles automatically from several sources, including headings, anchor text and og:title. It may also write its own snippet instead of using your description.
Reading Your Meta Tags Check: Six Problems and Their Fixes
When you check meta tags across a site, most failures fall into one of six patterns. Here's how each one looks, why it happens and what fixes it.
| Symptom | Likely cause | Fix |
|---|---|---|
| Same description on dozens of pages | Template default, never overridden | Set a per-page description field |
| Page drops out of search | noindex left over from staging, in the meta tag or X-Robots-Tag | Remove it from the template and the server config |
| Wrong URL ranks | Canonical points at a redirect, a noindexed page or another language | Point it at the final, indexable URL in the same language |
| Tags visible in DevTools only | A script injects them after load | Render them on the server or at build time |
| Tags ignored despite being in the source | Placed in the body, or an invalid element such as img earlier in the head | Move them above it and remove the invalid element |
| Old preview on social networks | Platform cache | Re-scrape in the debugger or Post Inspector |
The fifth row deserves a note. Google's guidance on valid page metadata says that when it meets an invalid element in the head, such as an img or iframe, it ignores every element after it. Other crawlers may recover differently, so keep the head valid and put your meta tags near the top.
We prefer fixing tags at the template or front matter level, not page by page. One template fix repairs every page that shares it. When you need replacements, write corrected tags with the meta tag generator. It runs in your browser, writes meta and Open Graph tags, and shows a search preview whose pixel widths are estimates. It doesn't fetch or crawl a URL, so you still check the live page with the methods above.
Fixing Meta Tags on a Markdown-Based Site
Markdown has no meta tag syntax. On a static site, fields in Markdown front matter such as title and description feed the site template, and the template writes the head. That's why a missing tag is usually a template bug, not a content bug. To add a meta tag to your website, set the field in your CMS or front matter, or paste the tag into the template's head.
If you're unsure where Markdown ends and HTML begins, Markdown vs HTML covers the split. A page you convert from Markdown to HTML gets whatever head the converter or template adds, so run view source on the output too.
Questions About Checking Meta Tags
The fastest way to check meta tags is to read all three layers: raw HTML, the rendered page and the response headers. Then confirm what Google and the social networks fetched. Fix problems in the template so every page benefits. For new tags, use the meta tag generator. You can draft the page copy in Markdown in the free editor, with no account needed.