Meta tags are the first thing Google and social networks read on your page. Four ways to check them, from zero-install to fully automated in CI.
Updated August 2026 · 5 min read
# See every meta tag on one page $ curl -s https://yoursite.com | grep -o '<meta[^>]*>' | head -20 # Just the title and description $ curl -s https://yoursite.com | grep -Eo '<title>.*</title>|name="description" content="[^"]*"'
| Tag | Rule of thumb |
|---|---|
| <title> | 30–60 characters, unique per page, most important words first |
| meta description | 120–158 characters, concrete message plus a reason to click |
| canonical | Exactly one per page, self-referencing (absolute URL) |
| og:title / og:description / og:image | Mirror title/description; image at least 1200x630 |
| robots | Only present when you want to discourage indexing |
Also read: broken link checkers, check redirect chains and technical SEO checks.
The title tag and meta description are the big two: they form the text in the search result. Canonical, hreflang and robots meta control indexing, while Open Graph and Twitter tags mainly affect how the page looks when shared.
Almost always missing or cached Open Graph data. Verify that og:image, og:title and og:description exist and that the image is at least 1200x630 px, then force a re-scrape in the sharing debugger after fixing.
Yes — right-click and choose View Page Source, or use curl. All meta tags live in the head section. It's fast for one page but impractical when auditing a whole site.
Run a script in CI that fetches each important page and validates that title, description and canonical exist with sensible lengths. That catches mistakes before release instead of after.
Related: Broken link checkers · Redirect chains · Technical SEO checks