OnlyAI SEO

Documentation

Titles, robots and social cards

Open OnlyAI SEO, Settings, Titles & social. This is where you decide how whole kinds of pages look in search results and when they are shared, and which pages search engines should leave out.

Nothing changes until you choose

A template left empty keeps what WordPress and your theme do today. A post's own SEO title and description (in the editor) always win over a template. The one default that does something: date archives are set to noindex, because they only repeat your posts.

If Yoast SEO, Rank Math, All in One SEO, SEOPress or The SEO Framework is active, OnlyAI SEO prints none of these tags, so you never get two of anything. Switch the other plugin off after you import from it.

Templates

There is one template for the home page, one per content type (posts, pages and any custom type), one per category, tag or other archive, and one each for author archives, date archives, search results and the 404 page. Each can have a title and a description. For example:

%title% %sep% %sitename%
%term% archive %sep% %sitename%
Results for %searchterm% %sep% %sitename%
%title%The post or page title
%sitename%Your site name
%sep%The separator you chose (or your theme's)
%tagline%Your site's tagline
%year%The current year
%excerpt%The post excerpt, or its first words
%category%The post's first category
%author%The author's display name
%date%The post date
%term%The category, tag or other archive name
%term_desc%That archive's description
%pt% / %pt_plural%The content type, singular or plural
%searchterm%What a visitor searched for
%page%"Page 2" on page two of an archive, empty otherwise

Categories, tags and other archives

On each term's edit screen there are three extra fields: an SEO title, a meta description and a "hide from search results" box. They win over the archive template.

Hiding pages from search (robots)

  • One page. In the editor, the Search & social panel has switches for noindex, nofollow, no cached copy and no text snippet, and a field for a canonical URL.
  • A whole kind of page. Tick "Hide all of these from search results" for a content type or taxonomy, or noindex author and date archives.
  • The sitemap follows. Anything marked noindex is left out of the XML sitemap, so you do not send search engines mixed signals.
  • Canonical URL. Only set one when the same content really lives at another address you want search engines to prefer. A wrong canonical can make a page disappear from results.

The robots.txt file

Open OnlyAI SEO, Settings, robots.txt to see the file WordPress serves and, if you need to, replace it. Nothing changes until you press Check and save, and a file is only saved after it has been read the way Google reads it (the most specific crawler group applies, the longest matching rule wins, Allow beats Disallow when they tie, and * and $ work in paths) and tested against your home page, your sitemap, your latest posts, your theme stylesheet, a WordPress script, your uploaded images and admin-ajax.php.

  • Refused. A rule before any User-agent line, a misspelt directive such as Dissallow (crawlers skip it without telling you), a line that is not a rule, a Sitemap address that is not a full address, an empty file, or a file over 500 KB.
  • Needs your confirmation. A file that would block your home page, your sitemap or your latest posts for Google or Bing. The list names each address and the rule that blocks it. Tick the box only on a site you really want hidden, such as a staging copy.
  • Warned. Blocked CSS, scripts or images (Google needs them to render the page), a blocked admin-ajax.php, Crawl-delay and Noindex: (Google ignores both), and paths that do not start with a slash.
  • Also here. Test an address shows whether any address is allowed for Googlebot, Bingbot, GPTBot or other crawlers and which line decided it. Load the WordPress default gives you a starting point,Restore the previous version undoes the last save, and Use the WordPress default switches the editor off.
  • Good to know. The sitemap line and your AI-crawler rules (OnlyAI SEO, AI Visibility) are still added after your text, and the check includes them. If a real robots.txt file sits in your site folder, or "Discourage search engines" is ticked in Settings, Reading, the editor tells you it is being overridden.

More than one focus keyword

Under the focus keyword in the editor, More keywords takes up to four more phrases the page should rank for (type one and press Enter). Each gets its own checklist: is it in the SEO title, the meta description, the URL, the first 100 words and a subheading, and is it used a sensible number of times. The count next to each phrase shows how many checks pass. These checks are guidance: the page score keeps following the main keyword, so adding phrases never lowers it. The extra keywords are also in Quick Edit, shown as +N in the Focus keyword column, and are imported from Rank Math, SEOPress, Yoast SEO Premium and All in One SEO.

Same keyword on two posts. If the main or an extra keyword is already the main or an extra keyword of another post, the editor says so under the keyword and links to that post. It is only a warning: two pages aimed at one phrase can compete in search results, so aim them at different phrases or merge them. Drafts and private posts you cannot open are never named.

WooCommerce

With WooCommerce active, OnlyAI SEO, Settings, WooCommerce has three switches, all on by default, and nothing changes without WooCommerce:

  • Cart, Checkout and My account are marked noindex and left out of the XML sitemap. They are different for every visitor.
  • Sorted and filtered shop pages (?orderby=, ?min_price=, ?filter_colour= and similar) are noindex while their links are still followed, so the original page is the one that ranks. Page 2 of a shop is not treated as a filter.
  • Products share as products. The Open Graph card gets the price, currency, availability and SKU.

Product breadcrumbs (Home, Shop, category, sub-category, product) work on their own, and category pages get Shop in front. WooCommerce prints its own Product structured data, so a Product block from the Rich result panel is not added on a product page (Review, Course and the others still are).

Social cards

When someone shares a link on Facebook, LinkedIn, X or a messenger, the card comes from Open Graph and X (Twitter) tags. OnlyAI SEO prints them on posts, pages, the home page, category and tag archives and author pages. Per post you can set a share title, description and image in the editor; otherwise the SEO title, the meta description and the featured image are used, then your site's default image. Add your X username and Facebook page in the same settings card. The best image size is 1200 by 630 pixels.

Archive hygiene

  • Empty archives. A category, tag or other archive with no posts is a thin page. Tick Hide empty archives and they are marked noindex while they stay empty.
  • Turn off date, author or post-format archives. Visitors and search engines are sent to the home page with a permanent (301) redirect. Turn off author archives only on a one-author site, or if nobody browses by author.

Working from the post list

In Posts and Pages, open Screen Options and tick SEO title, Meta description, Focus keyword and Links. Each shows the value, the length of titles and descriptions (red when too long) and, for links, how many posts link to this one, how many of your own pages it links to and how many outside links it has, with a warning on posts nothing links to. The page draws first and the incoming count is filled in a moment later from one link index that reads each post once and refreshes at most every six hours, so the list stays quick on sites with thousands of posts. Quick Edit now has the SEO title, description, focus keyword, "hide from search results" and "cornerstone content". Bulk Edit can hide or show many posts in search and mark or unmark them as cornerstone.

An HTML sitemap for visitors

Put [onlyai_sitemap] on a page to list your pages and posts (posts grouped by their main category). Options: types="page,post", group="none", limit="200", order="date". Pages hidden from search, password-protected pages and drafts are left out, and the list refreshes whenever a post changes. It helps readers, and gives crawlers a link to every page.

Image attachment pages and the RSS feed

  • Attachment pages (off by default). WordPress gives every uploaded image its own page, which is thin content. Turn this on to send those pages to the post the image belongs to, or to the file itself when it has no post.
  • RSS credit line (on by default). Every feed item ends with "The post [title] appeared first on [your site]". Sites that copy your feed then link back to you.