Meta Tags Checker
Inspect pasted HTML metadata for title, description, canonical, robots, Open Graph, X cards, duplicates, and placement.
Add the URL the HTML belongs to if you want relative canonical/social URLs resolved and the canonical compared with the inspected page. Nothing is fetched.
Paste the HTML source you want to inspect. Parsing stays in the browser. No live page, HTTP headers, robots.txt, sitemap, JavaScript-rendered network responses, or social image are fetched.
Search, canonical, robots, charset, social, duplicate, and placement findings will appear here.
Conflicting Signals Matter More Than a Checklist Score
A page can have a perfectly reasonable title and description while also carrying two canonical tags, a noindex directive, an Open Graph URL with campaign parameters, or metadata injected outside the document head. A simple checklist that says “title present, description present” can miss the part that actually changes how the page is interpreted.
Individual fields are reported first, followed by relationship and placement checks where they change the meaning. It does not turn the result into a percentage score because metadata quality is not a point system.
The Supplied Page URL Turns a Relative Canonical From a String Into a Relationship
Page being inspected: https://example.com/docs/page HTML: <link rel="canonical" href="/docs/page"> Resolved canonical: https://example.com/docs/page
Without the page URL, a relative value can only be described as relative. With the page URL, it can be resolved, compared with the inspected URL, and checked for problems such as an HTTPS page canonically pointing to HTTP.
Absolute canonical URLs remain easier to audit and less vulnerable to base-URL mistakes, which is why Google recommends them as a safer implementation choice.
Canonical and noindex Are Not Two Ways to Say “Do Not Show This URL”
A canonical is a hint about which URL should represent duplicate or very similar content. noindex is an indexing directive telling a supporting crawler not to show that page in search results. They solve different problems.
Google can choose a different canonical from the one you specify, while a page carrying a usable noindex directive is asking not to be indexed. Review combinations deliberately instead of treating both as generic “SEO tags.”
Campaign Parameters Usually Describe the Visit, Not the Canonical Document
URLs containing utm_source, gclid, fbclid, or similar identifiers are often tracking variants of a stable content URL. If one of those variants appears in rel=canonical, review whether the page is accidentally declaring the campaign-specific URL as its representative.
A warning appears rather than deleting parameters automatically. Some applications genuinely use query parameters to identify distinct content, so canonicalization needs site context.
Search Title and Description Lengths Are Display Constraints, Not Fixed Ranking Rules
Search engines can rewrite title links and build snippets from page content. Google does not publish a fixed title character limit or a fixed meta-description character limit that guarantees how many characters appear in every result.
Character counts still provide editorial context: a title can be obviously bloated or a description can be empty. The important distinction is not to convert those counts into invented pass/fail SEO rules.
Character Encoding Is One Metadata Field Where Physical Source Position Matters
Current HTML authoring rules allow a meta-based encoding declaration through charset or the legacy http-equivencoding form, require that declaration to identify UTF-8, and require the complete declaration within the first 1024 bytes. Source position therefore matters in addition to what DOMParser recovers.
The HTTP Content-Type header or a byte-order mark can also affect encoding detection. Pasted text has already been decoded by the browser, so the byte-position check uses its UTF-8 re-serialization and cannot reconstruct the original response bytes.
Open Graph and X Metadata Can Be Correct While the Preview Is Still Wrong
A valid og:imagestring does not prove that the image exists, returns the expected content type, is reachable by the social crawler, fits the platform's image requirements, or has refreshed in that platform's preview cache.
Markup inspection stops at the declaration. Verify a fetched card with the platform's own preview or debugging workflow when image reachability, cache state, or rendered card behavior matters.
Metadata Outside <head> Is Worth Investigating Even If DOMParser Can Still See It
Browser HTML parsing can recover from malformed markup, so DOM queries may still find metadata placed somewhere unexpected. Google currently documents that robots meta rules can be respected in the body, but that exception does not make arbitrary metadata placement a sound authoring pattern. Canonical links, social metadata, charset declarations, and other head content have their own requirements.
Inspect the deployed HTML when duplicate layouts, nested head managers, hydration, or CMS plugins may be generating tags in more than one place.
Why Meta Keywords Are Only Reported When They Exist
Google Search does not use the meta keywords tag for web ranking. Adding a “keywords” field to every metadata checklist can look comprehensive while teaching an obsolete workflow.
If a pasted page already contains the tag, Yoryantra reports it so you can decide whether some non-Google system still depends on it. Its absence is not a problem that needs fixing.