cd /news/ai-agents/structured-data-json-ld · home topics ai-agents article
[ARTICLE · art-31107] src=specification.website ↗ pub= topic=ai-agents verified=true sentiment=· neutral

Structured data (JSON-LD)

JSON-LD structured data using schema.org vocabulary is the standard for machine-readable page annotations, enabling search engines to display rich results and AI agents to extract factual ground truth. Implementation requires mirroring visible content, using absolute URLs, and validating with tools like Schema.org and Google's Rich Results Test.

read3 min views1 publishedJun 10, 2026

Machine-readable annotations that describe the content of a page using the schema.org vocabulary. JSON-LD is the format search engines and AI agents expect.

What it is #

Structured data is a set of machine-readable statements that describe what a page is about, using the shared vocabulary at schema.org. The recommended serialisation is JSON-LD: a <script type="application/ld+json">

block inside <head>

(or, less commonly, <body>

).

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "Article",
  "headline": "What is HSTS?",
  "datePublished": "2026-05-29",
  "author": {
    "@type": "Person",
    "name": "Jane Doe"
  },
  "publisher": {
    "@type": "Organization",
    "name": "Example",
    "url": "https://example.com"
  }
}
</script>

Microdata and RDFa are also accepted, but JSON-LD is the de facto standard because it sits separate from the visible markup.

Why it matters #

Two audiences read it heavily:

Search engines use structured data to power rich results (article cards, breadcrumbs, product carousels, knowledge-panel facts). Without it, you get a plain blue link.AI agents and answer engines rely on it as the ground truth for facts they may quote. APerson

schema with asameAs

linking to your verified profiles is the cleanest way to assert identity.

It is also the most stable contract between a publisher and the rest of the web. The HTML can change; the JSON-LD describes meaning.

How to implement #

Stick to a small set of well-supported types:

— site-wide, on the home page. IncludeWebSite

url

andname

, andpotentialAction

for sitelinks search if appropriate.orOrganization

— for the publisher and authors. IncludePerson

sameAs

arrays pointing at verified profiles.— on every page that has a breadcrumb trail.BreadcrumbList

orArticle

— for articles, withBlogPosting

headline

,datePublished

,dateModified

,author

,image

.,Product

,Offer

— for e-commerce, where eligibility is strict.AggregateRating

— only when the page genuinely has a Q-and-A list visible to users. Note that GoogleFAQPage

retired the FAQ rich result in 2026:FAQPage

is still valid schema.org vocabulary, but it no longer produces a Google search feature, and no answer engine has confirmed it favours the markup over rendered HTML — so add it for genuine visible content, not for SERP or “GEO” gain. Never stuff fake FAQs. (Further readingon why the format was abused into deprecation, and the proposedFAQSection

type for Q-and-A that is asectionof a page rather than its main entity.)

Rules:

Mirror what is visible on the page. Do not declare facts in JSON-LD that the page does not state. Google calls this “out of sync” data and ignores or penalises it.One graph per page is cleaner than many fragments. Use@graph

to nest related entities and@id

URIs to cross-reference them.Use absolute URLs in@id

,url

,image

, andsameAs

.**Keep dates in ISO 8601.**Validate. Schema.org evolves; what is valid one year may be deprecated the next.

Common mistakes #

  • Fabricating aggregateRating

orReview

to win stars. Google detects this and removes the rich result, sometimes the whole site’s eligibility. - Marking up navigation, footers, or sidebars as if they were the main content.

  • Forgetting to update structured data when the page content changes.
  • Multiple disagreeing @type

declarations across plugins or templates on the same page.

Verification #

Sources & further reading #

Schema.org— schema.org - FAQPage (FAQ) structured data— Google Search Central - Introduction to structured data markup in Google Search— Google Search Central - JSON-LD 1.1 Specification— W3C

── more in #ai-agents 4 stories · sorted by recency
── more on @schema.org 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/structured-data-json…] indexed:0 read:3min 2026-06-10 ·