Home/Tools/AI-Ready Website Checklist

ENGINEERING & ARCHITECTURE SPECIFICATION

AI-Ready Website Checklist.

A rigorous pre-launch engineering checklist for developers, software architects, and technical leads building modern websites structured for AI retrieval, answer engines, and autonomous agents.

Project / Repository Under Audit:

Review Date: 27 August 2026

35%

Tech Readiness

Legacy Web Architecture

7 / 20 verified

DOM & Semantic HTML5 · tech-01

Strict Semantic HTML5 Structure (<main>, <article>, <section>)

Replace generic `<div>` soup with landmark HTML5 elements. Ensure the primary page content lives exclusively within `<main>` and discrete articles within `<article>`.

<main id="main">
  <article className="prose">
    <h1>Direct Topic Headline</h1>
  </article>
</main>

Why It's Critical:Headless LLM scrapers extract text by traversing standard HTML5 document landmarks.

Verification Method:Run browser accessibility tree inspect or W3C HTML validator.

DOM & Semantic HTML5 · tech-02

Hierarchical Heading Architecture (H1 -> H2 -> H3)

Exactly one `<h1>` per page matching the document entity. Section headers strictly nested without skipping levels (e.g. no jumping from H1 directly to H4).

Why It's Critical:Generative search engines parse section hierarchy to determine sub-topic parentage.

Verification Method:Audit document outline via HeadingsMap extension or Cheerio parser.

DOM & Semantic HTML5 · tech-03

Tabular Data in Semantic <table> Elements

Structure comparative data, price matrices, and specifications using semantic `<table>`, `<thead>`, `<th>`, and `<tbody>` rather than CSS flex/grid divs.

<table>
  <thead><tr><th>Tier</th><th>Deliverable</th></tr></thead>
  <tbody><tr><td>DIP Beacon</td><td>AI Tracking</td></tr></tbody>
</table>

Why It's Critical:LLMs parse tabular matrix tokens with significantly higher structural accuracy than freeform styled divs.

Verification Method:Confirm table structure contains header rows with `scope="col"`.

DOM & Semantic HTML5 · tech-04

Descriptive Microcopy & Accessible Image Alt Attributes

All technical diagrams, workflow graphics, and UI previews must have complete, context-rich `alt="..."` descriptions and caption text.

Why It's Critical:Multimodal and text-only LLMs rely on alt attributes to index visual workflows without OCR latency.

Verification Method:Audit page with Lighthouse Accessibility or axe DevTools.

Schema Knowledge Graph · tech-05

Top-Level @graph Array Unification

Unify all isolated `<script type="application/ld+json">` tags into a single root `@graph` network connecting Organization, WebSite, WebPage, and BreadcrumbList.

{
  "@context": "https://schema.org",
  "@graph": [
    { "@type": "Organization", "@id": "https://example.com/#org" },
    { "@type": "WebSite", "@id": "https://example.com/#site", "publisher": { "@id": "https://example.com/#org" } }
  ]
}

Why It's Critical:Disconnected single-entity tags create fragmented entity graphs that LLMs cannot resolve to parent organizations.

Verification Method:Test via Schema.org Validator or Google Rich Results Test.

Schema Knowledge Graph · tech-06

Explicit @id URI Anchoring on All Entities

Assign unambiguous `#entity` fragment identifiers to all core nodes (e.g. `https://example.com/#organization`, `https://example.com/#primary-clinic`).

Why It's Critical:Enables deterministic cross-referencing between pages without duplicating entity payload blobs.

Verification Method:Verify `@id` strings are valid absolute URIs with fragment identifiers.

Schema Knowledge Graph · tech-07

Comprehensive sameAs Cross-Verification Nodes

Include verified links to official corporate registries (ACRA, Singapore government directories, LinkedIn company pages, Google Maps CID).

Why It's Critical:AI models cross-corroborate corporate claims against third-party ground-truth datasets.

Verification Method:Ensure `sameAs` array contains active HTTPS links to verified public registries.

Schema Knowledge Graph · tech-08

ISO 8601 Date Formatting & dateModified Headers

Use complete ISO 8601 format (`YYYY-MM-DD` or `YYYY-MM-DDTHH:MM:SS+08:00`) for `datePublished` and `dateModified` in Article/TechArticle schema.

Why It's Critical:LLMs require strict temporal anchors to assess information freshness and avoid serving deprecated specs.

Verification Method:Confirm date strings parse through `new Date(string).toISOString()`.

Crawler & SSR Infrastructure · tech-09

Server-Side Rendering (SSR) / Static Site Generation (SSG)

Deliver pre-rendered HTML on first byte. Avoid client-only client rendering (CSR) where content is injected after JavaScript execution.

Why It's Critical:Headless AI crawlers (GPTBot, ClaudeBot) operate strict execution timeouts and often skip heavy client hydration.

Verification Method:Disable JavaScript in browser devtools and confirm full text renders on initial page load.

Crawler & SSR Infrastructure · tech-10

Time to First Byte (TTFB) < 300ms on Singapore Edge

Serve static assets and cached HTML via edge CDN (Vercel, Cloudflare) with TTFB consistently below 300ms from Singapore nodes.

Why It's Critical:Slow backend response times increase crawler drop rates during high-frequency real-time citation lookups.

Verification Method:Benchmark with WebPageTest from Singapore EC2 region.

Crawler & SSR Infrastructure · tech-11

Explicit AI Crawler Directives in robots.txt

Declare explicit `User-agent: GPTBot`, `ClaudeBot`, and `PerplexityBot` rules with `Allow: /` for indexable paths.

User-agent: GPTBot
Allow: /

User-agent: ClaudeBot
Allow: /

User-agent: PerplexityBot
Allow: /

Why It's Critical:Prevent ambiguous fallback to restrictive global Disallow rules.

Verification Method:Validate using Infinitus AI Crawler Checker.

Crawler & SSR Infrastructure · tech-12

Automated XML Sitemap with Priority & lastmod Flags

Maintain an automated `sitemap.xml` generated at build time with valid `<lastmod>` timestamps matching content git commit history.

Why It's Critical:Guides search and AI discovery engines to freshly updated procedural playbooks and fee tables.

Verification Method:Check XML sitemap response header `Content-Type: application/xml`.

Answer Density & Content Chunking · tech-13

Direct Answer In First 30% of DOM

Place direct, factual summaries (40–60 words) immediately beneath primary topic H2 headings before explanatory background.

Why It's Critical:Vector embedding search models give highest retrieval weighting to the first chunk of a semantic section.

Verification Method:Verify key definition paragraph is visible above the fold on mobile viewports.

Answer Density & Content Chunking · tech-14

Factual Claim Backing with Citation Sources

Ground statistical claims, regulatory rules, and clinical facts with linked source citations to government registries (MOH, HSA, ACRA) or RFCs.

Why It's Critical:AI citation extractors favor texts containing verifiable external citation anchors.

Verification Method:Audit external links for valid DOIs, RFC standards, or `.gov.sg` domains.

Answer Density & Content Chunking · tech-15

Deterministic FAQ Schema Pairing with On-Page Q&A

Ensure every question-answer pair listed in on-page accordions has an exact 1:1 match inside the `FAQPage` schema `mainEntity` array.

Why It's Critical:Mismatches between schema text and rendered HTML trigger search quality penalty flags.

Verification Method:Compare text nodes in DOM against JSON-LD script payload.

Answer Density & Content Chunking · tech-16

Mobile Touch Targets & Core Web Vitals Compliance

Zero layout shift (CLS < 0.05), Largest Contentful Paint (LCP < 2.0s), and touch target minimums (48x48px) on mobile viewports.

Why It's Critical:Google AI Overviews and mobile-first search engines enforce CWV thresholds as quality gates.

Verification Method:Run Google PageSpeed Insights mobile audit (Score >= 90).

Machine Endpoints & Future Context · tech-17

Root /llms.txt Machine Context File

Deploy a root-level `/llms.txt` markdown file providing LLMs with concise company context, key service endpoints, and documentation under 2,500 tokens.

Why It's Critical:Enables emerging autonomous AI search agents to parse website context with minimal token overhead.

Verification Method:Validate using Infinitus llms.txt Validator.

Machine Endpoints & Future Context · tech-18

Secondary /llms-full.txt Extended Dataset

Deploy an extended `/llms-full.txt` containing full research transcripts, case study datasets, and technical documentation.

Why It's Critical:Provides full-depth context for agentic reasoning workflows without bloating the root token footprint.

Verification Method:Confirm `/llms-full.txt` is referenced as an optional link in `/llms.txt`.

Machine Endpoints & Future Context · tech-19

Machine-Readable Open Graph & Twitter Meta Tags

Deploy complete `og:title`, `og:description`, `og:image` (1200x630px), `og:locale` (`en_SG`), and `twitter:card` tags on every page.

Why It's Critical:Used by social aggregators and preview bots to extract metadata for citation cards.

Verification Method:Inspect HTML `<head>` or use OpenGraph preview debugger.

Machine Endpoints & Future Context · tech-20

Deterministic Error Handlers & 404 Header Validation

Custom 404 page must return HTTP status 404 Not Found (never HTTP 200 soft-404) and link back to core site navigation.

Why It's Critical:Soft-404 pages corrupt LLM index caches by assigning indexable status to non-existent URLs.

Verification Method:Test non-existent URL via cURL: `curl -I https://example.com/nonexistent` returns HTTP 404.

Need Help Implementing This Technical Architecture?

Infinitus works directly with engineering teams and enterprise brands to build end-to-end JSON-LD knowledge graphs, SSR infrastructure, and machine discovery endpoints.

Explore AI Visibility Services →