Docs / for WordPress / FAQ

AEO Ultimate for WordPress

Questions people actually ask

If your question is about something going wrong, Troubleshooting is the better page. This one is about how the plugin behaves and why.

Which schema types does this plugin generate?

JSON-LD for Organization, Article, NewsArticle, Person and author (with EducationalOccupationalCredential and sameAs), FAQPage, Service, Product (with Brand and Offer), LocalBusiness (with opening hours and geo coordinates), ContactPoint, Event, Review and AggregateRating, and BreadcrumbList.

Each type is added only when it suits the page and is not already present. That second condition is why you may see fewer types than this list suggests — if another plugin already published something correct, this one does not duplicate it.

Does this conflict with Yoast SEO, Rank Math or AIOSEO?

No, and it is designed around not conflicting. The plugin detects active SEO suites and steps back from any metadata they already output — Organization, Article, Open Graph. It acts as an enhancement layer, adding the signals those suites typically miss: article:publisher, rate-limit controls for 20+ AI agents, author credential schema and llms.txt processing.

Most people run both. That is the intended setup, not a compromise.

What is llms.txt and why would I need it?

llms.txt is an emerging standard (see llmstxt.org) that gives AI systems a clean, machine-readable summary of what your site contains and how to reach it — think of it as a robots.txt written for language models.

The companion endpoint /llms-full.txt serves your content as clean Markdown, so a crawler can ingest it without parsing token-heavy HTML. Both are served virtually — no files are written to your site.

Will this slow my site down?

No. Front-end metadata, schema and virtual assets are served from the Transients API with a one-hour cache. Heavier work runs only when a post is saved or settings change. The crawler rate limiter adds a single transient read per page load, and only for known AI agents — ordinary visitors never touch that path.

Will blocking AI bots affect my human visitors or Google?

No. The filter matches requests by User-Agent against the known AI crawler list only. Regular browsers, ordinary visitors and traditional search engines such as Googlebot pass through unaffected.

Blocking a bot does two things at once: it writes a Disallow entry to your virtual robots.txt and returns a 403 to that agent. Advisory and enforced together, because the advisory half alone is widely ignored.

Why does the publisher and author split matter?

Standard social tags often blur authorship by failing to separate the brand from the individual writer. This plugin keeps them distinct: article:publisher establishes the company identity, article:author links to the person. AI models use that split to assess authority and attribute content accurately — a page that cannot say who wrote it is harder to cite.

What is the OKF bundle?

Google’s Open Knowledge Format (v0.2) publishes your organisation, products, categories, posts and policies as Markdown concept files that AI systems can ingest. It is served at /okf/ automatically — nothing to configure.

Which AI agent-discovery standards are supported?

An Agent Skills Index at /.well-known/agent-skills/, an MCP server card with WebMCP context injection, OAuth/OIDC discovery, an RFC 9727 API catalog, RFC 8288 Link headers, Content-Signal headers and semantic breadcrumb signals — alongside llms.txt, llms-full.txt and Markdown content negotiation.

Each toggles independently. Where automatic /.well-known/ file creation needs server write access you do not have, you get copy-and-paste instructions instead of a silent failure.

What does Meta Business Sync do?

It adds the sitewide fb:app_id tag, linking your domain to your Meta Developer App. That unlocks Meta Domain Insights and helps Meta attribute your content correctly across its ad and AI systems. Most SEO plugins do not output this tag at all.

Does it monitor performance and analytics?

Yes, through the Command Center: Google Search Console, Google Analytics 4 and Bing Webmaster Tools in one view, plus PageSpeed Insights audits — the Lighthouse score and Core Web Vitals (LCP, CLS, FCP, TBT), cached for six hours.

A traffic-leak scan combines GA4 and PageSpeed to surface high-traffic pages that bounce hard on slow mobile loads. Everything is optional and stays off until you connect the service with your own credentials.

What are the system requirements?

Which external services does it connect to?

Only the ones you configure, and only when you trigger them. AI providers — Anthropic, OpenAI, Google Gemini, Perplexity, xAI and Mistral — each use a key you supply and are billed to your own account at the provider’s rate, with nothing added on top and no metered credits.

Nothing is sent anywhere until you configure a feature and trigger it. See the privacy policy for what leaves your server and when.

Something is not answered here

Ask. Questions that reach me become documentation, so asking one genuinely helps the next person with the same problem.