JSON-LD (JavaScript Object Notation for Linked Data) is a method of encoding linked data using JSON. It allows websites to inject structured semantic context into HTML without altering the visual page.

Why is JSON-LD important for AI agents?

Historically, JSON-LD was used for SEO rich snippets. Today, AI agents use it to understand web pages deterministically. wmcp.sh can ingest JSON-LD schemas and convert them into callable agent tools instantly.

What are the most common schema.org types?

The most common vocabulary schemas include Article, Product, Recipe, and FAQPage. These define standard properties like price, author, or step-by-step instructions.

Can JSON-LD power real-time interactions?

Yes, when coupled with a sub-100ms gateway like wmcp.sh, JSON-LD endpoints can serve as dynamic configuration files that update model context and available tools in real-time.

GLOSSARY · /GLOSSARY/JSON-LD

JSON-LD (Linked Data)

AI agents struggle to parse unstructured HTML. JSON-LD allows you to embed machine-readable semantics directly into your pages. wmcp.sh leverages this structured data to instantly convert any webpage into a callable tool.

the gap

Bridging the gap between content and execution

Originally designed for Google SEO rich snippets, the schema.org vocabulary provides standard structures for things like Product, Article, Recipe, and FAQPage. Traditionally, these were static declarations.

However, with the rise of autonomous agents, JSON-LD has a new purpose. Instead of merely indexing content, an AI model can read a JSON-LD payload to understand the exact structure of a page, its price variables, or available actions. wmcp.sh acts as a bridge, parsing JSON-LD on the fly and serving it back to agents as executable, sub-100ms tool schemas. (Note: wmcp.sh is not affiliated with Google or schema.org.)

the implementation

How to structure JSON-LD for AI extraction

<!-- Example embedded in the head of your HTML -->
<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "Product",
  "name": "Acme Corp Enterprise License",
  "description": "Software license for support@example.com environments.",
  "offers": {
    "@type": "Offer",
    "price": "999.00",
    "priceCurrency": "USD",
    "availability": "https://schema.org/InStock"
  }
}
</script>
<!-- wmcp.sh automatically extracts this block via its agent API -->

capability

Why JSON-LD matters for agents

Capability	Without Structured Data	With JSON-LD & wmcp.sh
Content Extraction	⚠️ Requires brittle CSS selector scraping.	✅ Perfect determinism via schema.org.
Execution Latency	❌ Scraping HTML DOMs takes 1-3 seconds.	✅ Extracted at the edge in sub-100ms.
Dynamic State	❌ Stale agent context.	✅ Micro-cached (short TTL, ~1s) payloads.
Authentication Requirements	⚠️ Hard to pass tokens to scraping bots.	✅ Processed via encrypted credentials vault.

FAQ

Common questions.

Does JSON-LD replace OpenAPI?

No. JSON-LD is excellent for descriptive entities (like a Product catalog or a Recipe), while OpenAPI is required for complex state mutations (like POST requests). wmcp.sh supports both paradigms interchangeably.

What is schema.org?

It is a collaborative, community activity whose mission is to create, maintain, and promote schemas for structured data on the Internet. It is the vocabulary most commonly used within JSON-LD payloads.

How do agents use FAQPage schema?

By extracting FAQPage JSON-LD, an agent can instantly inject vetted Q&A pairs into its context window, preventing hallucinations and reducing the need to hallucinate answers based on unstructured text.

Need this done for you?

Skip the wiring — we build, deploy, and monitor.

Custom adapter + hosted MCP at mcp.yourbrand.com + verified badge. Pricing: Starter $499 one-time, Managed Retainer $999/mo, or Enterprise $4,999+/mo.

See /managed → Submit (free)