site.url-variant-duplicate

What to change

A finding of this rule on a page carries the exact edits for that page: the markup it quotes, and what to write instead.

When it runs

Runs on a crawl or a build directory. A warning is a problem that costs a page ranking, clicks or citations. Each finding takes 5 points off the page's score.

The severity here is the rule's usual one: a finding can be raised or lowered when the page gives a reason.

Run it, or skip it

Check a page for this rule alone, after a fix for instance:

npx @rivi-software/agentic-seo-check crawl https://example.com --rule site.url-variant-duplicate

Or from an assistant, with the MCP tool seo_crawl_site:

{
  "url": "https://example.com",
  "rules": [
    "site.url-variant-duplicate"
  ]
}

To leave it out of an audit, pass --disable site.url-variant-duplicate on the command, or disableRules: ["site.url-variant-duplicate"] to a tool.

• Warningsite.canonical-target-noindexPages name as canonical a page that is set to noindex, so the page meant to be indexed is kept out of the index.
• Warningsite.canonical-chainA page's canonical points to a page whose own canonical points somewhere else (A to B to C).
• Infosite.links-to-noindexPages set to noindex are linked from other pages of the site. A noindex page that much of the site links to is often noindex by mistake. Cart, search, login and account pages are left out.
• Warningsite.hreflang-target-invalidhreflang alternates point to URLs that are broken, redirect, are noindex or name another URL as canonical, so the annotation is ignored.
• Warningsite.parameter-duplicatesCrawled URLs that differ only in their query string show the same text, and none names one of them as canonical.
• Warningsite.broken-imagesImages on the pages point to URLs that answer 4xx/5xx or do not load.
• Infosite.crawl-limit-reachedThe crawl stopped at the page limit, so site-wide checks (orphans, duplicates, sitemap coverage) are partial.
• Infosite.crawl-stoppedThe hosted server stopped requesting the site before the crawl was done: the site refused its requests (a challenge, HTTP 429), robots.txt asked for a Crawl-delay longer than a hosted crawl waits, or the server reached its request limit for the site. Site-wide checks (broken links, duplicates, sitemap coverage) are partial.