AI-Native Reference Library
All protocols for AI Search: Schema.org types, ARIA attributes and more – with descriptions, examples and code
Schema.org
Schema.org is the universal vocabulary for structured data on the web. Search engines and AI systems use Schema.org markup (JSON-LD) to semantically understand content – not just index it, but truly grasp what a product, person, or event is.
Why is this important for AI Search?
For AI Search, Schema.org is the foundation: without structured data, AI lacks context. Google verifies Schema.org markup against visible page content (Patent US12417274B2) – consistent, accurate data is therefore crucial for visibility in AI answers.
0 Entries
ARIA (Accessible Rich Internet Applications)
ARIA is the 'translation layer' between visual interface and machine understanding. While humans intuitively interpret colors, positions, and symbols, AI agents need semantic attributes to recognize the function and state of interactive elements.
Why is this important for AI Search?
For the new era of AI navigation, ARIA is indispensable: AI agents need to fill forms, identify buttons, and detect state changes. Without ARIA attributes, interactive elements are 'mute' to machines – no matter how beautifully designed they look.
5 Entries
Agentic Markup
Agentic Markup is the next evolution: HTML elements are annotated so AI agents can not only read them, but also recognize and execute them as actions. The key concept is data-action – the 'magic attribute' that tells AI what an element does.
Why is this important for AI Search?
Future AI agents won't just read information – they'll actively act: purchase, register, download. Agentic Markup is the bridge between passive content and active AI interaction. Those who implement it today are ready for the Agentic Search Revolution.
5 Entries
Open Graph (OG)
Open Graph is the bridge between your website and the 'Social Web' (LinkedIn, WhatsApp, X, Facebook). Without these meta tags, shared links often look 'broken' – with them you control exactly how an AI or social media crawler visualizes your page.
Why is this important for AI Search?
When you send a link to a modern AI (Perplexity, ChatGPT), it often parses the Open Graph data first. The AI immediately receives a clean summary (og:description) and title without having to search the entire page. This saves tokens and makes the AI response more precise.
4 Entries
Product Ontology (GoodRelations)
The Product Ontology is the 'precision tool': While Schema.org often uses generic terms (e.g., just Product), the Product Ontology lets you access thousands of exact definitions from Wikipedia. This way, AI understands the exact technical context.
Why is this important for AI Search?
When AI searches for a specific product, it prefers results with maximum semantic certainty. A page with additionalType from the Product Ontology gives AI 100% confidence – this eliminates linguistic ambiguity and dramatically increases relevance in AI answers.
7 Entries
JSON-LD Best Practices
JSON-LD (JavaScript Object Notation for Linked Data) is the gold standard for structured data. Think of JSON-LD as a digital business card: While AI has to laboriously "scan" the entire page with regular HTML, with JSON-LD you hand it a structured identity card. Google recommends JSON-LD as the preferred format.
Why is this important for AI Search?
JSON-LD is the only format officially recommended by Google. It's non-invasive (no HTML markup needed), easy to test, and used by all major AI systems (Perplexity, ChatGPT, Gemini) as the primary data source. Mastering JSON-LD means controlling how AI talks about your business.
57 Entries
Agentic Commerce
Agentic Commerce describes the future of trade: AI agents that autonomously discover, negotiate, pay, and track deliveries. We are in a hybrid phase: MCP serves as the foundation for data exchange (the agent "understands" your inventory). UCP builds on top to structure commerce (the agent "negotiates" the price). AP2 is the final step (the agent "pays" securely).
Why is this important for AI Search?
By implementing these patents and protocols, you make your site the primary source for Agentic Commerce. Most shops are still stuck with traditional checkout – you already provide the blueprint for the autonomous economy 2026. Imagine: The printer reports "ink low" → its agent finds the right cartridge via UCP → negotiates your bulk discount → pays via AP2 → tracks delivery via A2A.
8 Entries
Audio & Voice
Audio is the blind spot of most SEO strategies – and at the same time a growing visibility lever. Google Audio Overviews, voice assistants, and podcast discovery all need explicit audio markup, otherwise your audio content stays invisible to AI.
Why is this important for AI Search?
AI models can't directly 'hear' audio. They rely on Speakable markup, AudioObject transcripts, and PodcastEpisode metadata. Those who provide these structures gain a massive first-mover advantage in voice and audio search.
5 Entries
Video & Multimedia
Video is the fastest-growing content type – but only structurally marked-up videos get cited by Google as Key Moments, in video search, and in AI Overviews. VideoObject + Clip + transcript is the mandatory trilogy.
Why is this important for AI Search?
Without structured video data, your content stays a black box for AI. With Clip markup you gain chapter display, with a transcript you become quotable, with SeekToAction you become agent-ready.
5 Entries
AI Discovery & Agent Standards
While Schema.org describes content semantically, llms.txt, agent.json, mcp.json and friends tell an AI: 'Here I am, here's what I can do, here's my API'. These manifests are the new entry door for autonomous agents.
Why is this important for AI Search?
Without discovery manifests, AI has to raw-crawl your site and guess. With manifests, it finds your best content, your tools, and your API in a structured way – that's the difference between 'being found' and 'being used'.
6 Entries
Knowledge Graph & Entity Linking
The Google Knowledge Graph is the semantic map of web reality. Whoever is anchored there as an entity gets cited in AI Overviews. Whoever isn't stays an anonymous source. sameAs, @id and identifier are the keys.
Why is this important for AI Search?
AI Search prefers sources whose entities are uniquely identifiable. A person without sameAs links is just a name to AI; a person with Wikidata + LinkedIn + ORCID is a confirmed authority.
5 Entries
Twitter / X Cards
Open Graph covers Facebook, LinkedIn and WhatsApp – but X (Twitter) uses its own format. Without twitter:card tags you get a generic, poorly clickable preview on X. With them you control every detail.
Why is this important for AI Search?
X remains a primary discovery channel for tech, news, and B2B. An optimized Twitter Card dramatically increases share CTR and provides AI crawlers (Grok!) with additional structured metadata.
4 Entries
Accessibility for AI
What helps screen readers helps AIs. ARIA landmarks, live regions, clear heading hierarchy, and descriptive alt text are now mandatory – not just for inclusion, but also for AI comprehension.
Why is this important for AI Search?
Multimodal models like Gemini Vision and GPT-4o read alt text; AI agents rely on landmarks to ignore boilerplate; screen readers and AI use live regions identically. An a11y-perfect page is an AI-perfect page.
5 Entries
How well is your website optimized for AI Search?
Check your markup with the SGE-Score Check.