How do I set a canonical for a PDF or non-HTML file?

You can’t add a tag to a PDF, so send an HTTP "Link: ; rel="canonical"" response header. This tool generates that header for you.

Canonical Tag Generator — Free Online Canonical Tag Generator

How do I generate a canonical tag for a web page?

Open AnyTool’s Canonical Tag Generator, paste the preferred (canonical) URL of the page, and it instantly builds the <link rel="canonical" href="…" /> tag for your <head> plus an HTTP Link-header version for PDFs and other non-HTML files. Add any duplicate URLs and it generates the canonical tag each one should carry, all pointing at the preferred URL, while a validation panel flags relative URLs, http vs https, tracking parameters and trailing-slash issues. Everything runs in your browser — nothing is uploaded or crawled.

Builds an absolute, self-referencing rel="canonical" link tag for the page <head>
Also outputs the HTTP "Link: <…>; rel="canonical"" header for PDFs / non-HTML
Generates the matching canonical tag for each duplicate / variant URL
Live validation: relative URL, http not https, UTM params, trailing-slash and case mismatches
100% client-side — your URLs never leave the browser; a canonical is a hint, not a guarantee

What is

Canonical Tag (rel="canonical")

A canonical tag is a <link rel="canonical"> element in a page’s <head> (or an equivalent HTTP Link header) that tells search engines which URL is the preferred, master version of content available at several URLs. It consolidates duplicate-content signals onto one URL, but it is a hint, not a directive: Google may still choose a different canonical, and it neither removes a page from the index (use noindex) nor redirects users (use a 301).

SEO & Web

Related terms

Duplicate contentSelf-referencing canonical301 redirectnoindexHTTP Link header

Frequently Asked Questions

rel="canonical" is a suggestion to search engines, not a command. Google combines it with signals like internal links, sitemaps, redirects and HTTPS, and may select a different canonical than the one you declared. It does not remove a duplicate from the index by itself (that is what noindex does) and it is not a redirect (use a 301 to actually move users). Place one absolute, self-referencing canonical per page, point it at a URL that returns 200, and confirm the canonical Google actually chose using Search Console’s URL Inspection tool.

No. A canonical is a hint, not a directive. Google weighs it with other signals and can still pick a different canonical, so verify the chosen URL in Search Console.

Google recommends absolute URLs in rel="canonical" — for example https://example.com/page rather than /page. A relative path is resolved against the current page’s URL and can point at an unintended location, especially on parameterized or paginated URLs. This generator validates that your canonical is absolute and uses https, and warns when it is not.

Always absolute — the full URL with https:// and the host. Relative canonicals are fragile and can resolve to the wrong page.

Use rel="canonical" when the same or near-identical content sits on several URLs you want all kept accessible (URL parameters, http/https, trailing-slash variants, syndicated copies) and you want ranking signals consolidated onto one. Use noindex when a page should stay reachable for users but be kept out of search results, and never point a canonical at a noindex page. Use a 301 redirect when a page has permanently moved and both users and crawlers should be sent to the new URL.

Use a canonical for duplicate pages you want to keep live, noindex to remove a page from search entirely, and a 301 when you have actually moved a page.

Because non-HTML files such as PDFs have no <head>, the canonical is declared through an HTTP response header instead: Link: <https://example.com/file.pdf>; rel="canonical" (per RFC 5988). The Canonical Tag Generator outputs this header version alongside the HTML tag so you can configure it on your server or CDN for downloadable files.

You can’t add a <link> tag to a PDF, so send an HTTP "Link: <url>; rel="canonical"" response header. This tool generates that header for you.

Canonical Tag Generator

Generate a correct, absolute, self-referencing rel="canonical" link tag and the equivalent HTTP Link header from a preferred URL, plus a canonical tag for each duplicate URL, with built-in validation — entirely in your browser.

Features

Absolute, self-referencing <link rel="canonical"> tag for the page <head>
HTTP "Link: <…>; rel=canonical" header version for PDFs and other non-HTML files
Per-variant tags so every duplicate URL points to one preferred URL
Validation lint: relative URL, http vs https, tracking params, trailing-slash / case, fragments, cross-domain
100% client-side — URLs never uploaded; honest canonical-is-a-hint notice

PriceFree

CategorySEO & Web

PlatformAll (browser-based)

PrivacyURLs are processed locally and never uploaded or crawled

Input: No upload — paste the canonical URL and any duplicate URLs

Output: Copy the <link> tag, Copy the HTTP Link header, Copy per-variant tags

Detailed Explanation

How It Works

What the Canonical Tag Generator Produces

The Canonical Tag Generator turns a single preferred URL into a correct, absolute, self-referencing <link rel="canonical" href="…" /> element for a page’s <head>, plus the equivalent HTTP response header (Link: <url>; rel="canonical") for non-HTML files such as PDFs. Given a list of duplicate or variant URLs, it generates the canonical tag each one should carry, all pointing back at the preferred URL so search engines consolidate duplicate-content signals onto a single page.

Outputs both the HTML <link rel="canonical"> tag and the HTTP Link header (RFC 5988)
Uses absolute, https, self-referencing URLs by default — Google’s recommended form
Generates a matching canonical tag for every duplicate / variant URL supplied
Cross-domain canonicals are supported for syndicated content
All generation is pure client-side JavaScript — no server call, no CDN, no upload, no crawling

Methodology

How It Normalizes and Validates URLs

Each URL is parsed with the browser’s native URL API. The canonical href is optionally upgraded from http:// to https://, has tracking/UTM parameters and any fragment stripped, and is checked for being absolute. A lint pass then reports the common real-world canonical mistakes — relative URLs, http instead of https, UTM and other tracking parameters, trailing-slash and uppercase-in-path inconsistencies, URL fragments, a variant identical to the canonical, invalid duplicate lines, and cross-domain references — each with a concrete fix.

Parsing uses the standards-compliant URL constructor, not regex guessing
Optional enforce-https, strip-query and self-referencing toggles
Detects relative URLs, http vs https, UTM/tracking params, trailing-slash and case issues
Flags URL fragments (unsupported in canonicals) and variants equal to the canonical
Findings are ordered errors → warnings → notes → confirmations

Limitations

A Canonical Is a Hint, Not a Guarantee

rel="canonical" is a suggestion to search engines, not a directive: Google weighs it with other signals (internal links, sitemaps, redirects, HTTPS) and may still select a different canonical. A canonical does not remove a duplicate from the index by itself — that requires noindex — and it is not a redirect, so users still reach every URL; a 301 is needed to actually move a page. The recommended practice is one absolute, self-referencing canonical per page, pointed at a URL that returns 200 (never a redirected, 404 or noindex page), with the chosen canonical verified in Search Console’s URL Inspection.

Canonical is a hint — Google can choose a different canonical URL
It does not deindex a duplicate (use noindex) and is not a redirect (use 301)
Point the canonical at a 200 URL, not a redirect, 404 or noindex page
Use exactly one canonical per page, in <head> or via the HTTP header
Confirm the canonical Google selected with Search Console URL Inspection

Canonical tag generation: in-browser (AnyTool) vs typical online generators
Capability	AnyTool	Typical online generators
Processing	Runs entirely in your browser	Often server-side
Your URLs	Never uploaded or crawled, works offline	Sometimes sent to a server
HTTP Link header (for PDFs)	Generated alongside the tag	Rarely offered
Per-duplicate tags	A tag for each variant URL	Usually one tag only
Validation	Lint for relative/http/UTM/trailing-slash/case/cross-domain	Often none
Honesty about limits	Clear hint-not-a-guarantee notice	Sometimes implies a guarantee
Cost / signup	Free, no signup	Often gated or ad-heavy

AnyTool builds the canonical tag and HTTP header locally and uploads nothing. A canonical tag is a hint to search engines, not a guarantee — Google can still choose a different canonical; it does not deindex duplicates (use noindex) and is not a redirect (use a 301). Comparison as of June 2026.

Cite This Page

Title: Canonical Tag Generator (build a correct absolute self-referencing rel="canonical" link tag and HTTP Link header from a preferred URL, plus per-duplicate tags, with built-in validation — 100% client-side)
Author: AnyTool
Published: January 1, 2026
Modified: February 8, 2026
Section: SEO & Web
URL: https://www.anytool.tech/canonical-tag-generator

Formatted Citation

AnyTool. "Canonical Tag Generator (build a correct absolute self-referencing rel="canonical" link tag and HTTP Link header from a preferred URL, plus per-duplicate tags, with built-in validation — 100% client-side)." AnyTool, January 1, 2026. https://www.anytool.tech/canonical-tag-generator

A privacy-first, 100% client-side Canonical Tag Generator that turns a page’s preferred (canonical) URL into a correct rel="canonical" implementation entirely in the browser, with no server call, no CDN and no upload, so the URLs entered are never sent anywhere or crawled and the tool works offline once cached. From a single preferred URL it produces the standard HTML <link rel="canonical" href="…" /> element for the page <head> and the equivalent HTTP response-header version, Link: <https://example.com/page>; rel="canonical" (per RFC 5988), for non-HTML resources such as PDFs and images that cannot carry a <head> tag. Given an optional list of duplicate or variant URLs — for example the same page reached with UTM/tracking parameters, over http instead of https, with or without a trailing slash, or a syndicated copy on another domain — it generates the canonical tag each of those URLs should carry, all pointing back at the one preferred URL so that search engines consolidate duplicate-content signals onto a single version. Every URL is parsed with the browser’s native, standards-compliant URL constructor rather than regex guessing; the canonical href is, by user-toggleable options, upgraded from http:// to https://, stripped of tracking/UTM query parameters and of any #fragment (which Google does not support in canonicals), and made self-referencing by default — the form Google recommends. A built-in validation lint, updated live as the user types, surfaces the classic canonical mistakes with a concrete fix for each: a relative rather than absolute URL, http rather than https, the presence of common tracking parameters (utm_*, gclid, fbclid and more), trailing-slash and uppercase-in-path inconsistencies that create distinct URLs, unsupported URL fragments, a duplicate URL that is actually identical to the canonical (a self-reference to remove from the list), duplicate lines that are not valid absolute URLs, and cross-domain references (valid for syndication). The tool is deliberately and prominently HONEST about what a canonical can and cannot do: rel="canonical" is a HINT to search engines, not a directive, and Google weighs it against other signals (internal links, sitemaps, redirects, HTTPS) and can still choose a different canonical; it does NOT remove a duplicate from the index by itself — that requires a noindex robots directive — and it is NOT a redirect, so users still reach every URL and a 301 redirect is needed to actually move a page; the correct practice is exactly one absolute, self-referencing canonical per page, placed in <head> or sent via the HTTP header, pointed at a URL that returns HTTP 200 (never a redirected, 404 or noindex page), with the canonical Google actually selected verified in Google Search Console’s URL Inspection tool. It provides one-click copy for the link tag, the HTTP header and each per-variant tag, supports dark mode and mobile with a sticky copy bar, and adds no new dependency — all logic lives in a small pure client-side engine.