Schema map for AI agents

An XML sitemap lists your pages. A schema map lists your structured data. Instead of loading every page to read its JSON-LD, an AI agent or an NLWeb tool can read one file and get the schema of the whole site.

RankFlect SEO publishes the schema map at /schemamap.xml, with one feed per content type. Each feed holds the same JSON-LD that RankFlect SEO prints on your pages, so nothing new is written about your content. The format is the schema map format used for NLWeb and AI agents.

The schema map is free and off by default.

Turn it on

1. Go to RankFlect SEO → Settings → Sitemaps.
2. Find the Schema map for AI agents card. "All your structured data (JSON-LD) in one feed at /schemamap.xml, for AI agents and NLWeb. Noindexed, private and password-protected content is left out."
3. Turn on Publish the schema map.
4. Click Save changes.

A View schema map link appears on the card. The URL works right away, with no need to resave permalinks.

The schema map needs the Schema module. If it is off, the switch says "Turn on the Schema module first: the schema map lists the schema it prints on your pages." Turn it on under RankFlect SEO → Dashboard → Modules.

What gets published

URL	What it is
/schemamap.xml	The index: one entry per feed, with the date its newest post changed.
/schemamap/post.jsonl	The feed for posts. Pages are at /schemamap/page.jsonl, products at /schemamap/product.jsonl, and so on.
/schemamap/post/2.jsonl	The next 100 posts. Each feed holds up to 100 items, and larger content types get numbered feeds.

A feed is in JSON Lines format: one schema.org node per line, each with "@context": "https://schema.org", its @type and its @id. Nodes that many pages share, such as your organization and your website, are listed once per feed instead of once per post.

If your site uses plain permalinks (Settings → Permalinks → Plain), the URLs are /?rankflect_schemamap=index for the index and /?rankflect_schemamap=post&rankflect_schemamap_page=2 for a feed page. The View schema map link always uses the right one.

What is left out

The schema map follows the same rules as your XML sitemaps:

• Only published posts are included. Drafts, private posts and password-protected posts are left out.
• Posts set to noindex are left out.
• Content types set to noindex or switched off in the sitemap settings are left out.
• Attachment (media) pages are never included.
• While Settings → Reading → Discourage search engines from indexing this site is ticked, the schema map is not served at all.

The robots.txt line

While the schema map is on, RankFlect SEO adds a line to your virtual robots.txt, right under the Sitemap: line:

Schemamap: https://example.com/schemamap.xml

AI agents that look for a schema map find it there. If you already wrote a Schemamap: line in Extra rules, RankFlect SEO does not add a second one. See Edit robots.txt.

Updates and caching

The schema map and its feeds are cached for 12 hours. The cache is cleared when you publish, update, trash or delete a post, change its SEO fields, edit a category or tag, change RankFlect SEO settings or change your permalinks. A renamed post or a new noindex shows up in the feed on the next request.

Responses carry X-Robots-Tag: noindex, follow, so search engines do not list the feeds themselves in their results.

Turn it off

Turn off Publish the schema map and click Save changes. /schemamap.xml returns a 404 again and the Schemamap: line is removed from robots.txt. Your page schema does not change either way.

For developers

Five filters adjust the schema map:

Filter	Use
rankflect_schemamap_post_types	The content types listed in the schema map.
rankflect_schemamap_per_page	Items per feed page (100, up to 1,000).
rankflect_schemamap_cache_ttl	How long the schema map is cached, in seconds (12 hours).
rankflect_schemamap_changefreq	The changefreq value in the index (daily).
rankflect_schemamap_priority	The priority value in the index (0.8).

See Hooks and filters.

FAQ

Does Google use the schema map?

Google reads the structured data on your pages, as before. The schema map is for AI agents and tools that want your structured data in one place. There is no promise that any service will use it.

Is it the same as llms.txt?

No. llms.txt is a Markdown index of your key pages for AI tools. The schema map holds your structured data as JSON-LD. You can publish both.

Why do I get a 404 at /schemamap.xml?

The switch is off, the Schema module is off, or the site is set to discourage search engines. A page cache can also keep serving an old 404 for a while; clear it after you turn the schema map on.

Does it slow down my site?

No. Feeds are built only when someone asks for them, then cached. Your pages are not changed.