Nordax Digital Visibility
Architecture Specification

Table of Contents

1. 1. Security & Trust Signals
2. 2. Core Metadata Architecture
3. 3. Structured Data
4. 4. AI Discoverability Infrastructure
5. 5. Crawlability & Indexation
6. 6. Performance & Core Web Vitals
7. 7. Content Structure & Semantics
8. 8. Canonical Authority Control
9. 9. Knowledge Graph & Entity Signals
10. 10. AI Optimisation & Future Readiness
11. 11. NIST Alignment
12. 12. Deployment Validation
13. 13. Levels 2 & 3 Overview

Section 1

Security & Trust Signals

Establish trust, protect users, and improve ranking signals.

• X-Content-Type-Options set to nosniff
• X-Frame-Options set to DENY
• Referrer-Policy set to strict-origin-when-cross-origin
• Permissions-Policy configured to restrict camera, microphone, and geolocation access
• Strict-Transport-Security: max-age=63072000; includeSubDomains; preload
• X-DNS-Prefetch-Control enabled
• X-Powered-By header removed
• HTTPS enforced across entire domain
• HTTP traffic redirected to HTTPS
• Non-canonical domain versions redirected to canonical version

Section 2

Core Metadata Architecture

Ensure clear identity, authority, and proper indexing.

• Metadata base URL (metadataBase) defined
• Title template defined and consistent across all pages
• Unique, keyword-relevant title for each page
• Unique, descriptive meta description for each page
• Application name defined
• Publisher defined using legal entity name
• Creator defined using brand name
• Category defined appropriately
• Canonical URL defined for every page
• OpenGraph: title, description, URL, site name, images (1200×630 minimum), locale, content type
• Twitter cards: summary_large_image, title, description, image
• Apple web app title defined

Section 3

Structured Data Implementation

Enable machine-readable understanding of the business and content.

• Organization schema: legalName, name, url, logo, contactPoint, sameAs social links
• WebSite schema: name, url, publisher, language, description
• LocalBusiness schema (where applicable): address, phone, geo coordinates, hours, serviceArea
• Article schema on article pages: author, datePublished, dateModified, images, publisher
• BreadcrumbList schema on structured pages
• All structured data validated at validator.schema.org

• Service schema on service pages
• Person schema for founders and authors
• ImageObject schema for significant images (description, caption, creator)
• FAQPage schema on pages with genuine FAQ content
• isPartOf and hasPart schema relationships to demonstrate content depth

Note: FAQPage schema should only be applied where a genuine Q&A section exists. Google limits FAQ rich results to well-known authoritative sites.

• VideoObject schema on pages with video content
• AggregateRating / Review schema if verified reviews exist
• Product schema if selling products
• Speakable schema for voice assistant readiness (still in beta)

Section 4

AI Discoverability Infrastructure

Ensure AI systems can properly discover, interpret, and recommend the site.

robots.txt — AI Bot Policy

Explicitly allow the following user-agents in addition to the wildcard * rule:

• Sitemap directive present in robots.txt
• Host directive present with canonical domain

Note: GrokBot has been observed spoofing other user-agent strings. Allowlisting GrokBot and xAI-Grok ensures your intent is on record even if enforcement is imperfect.

• llms.txt present: business summary, site structure overview, contact information
• llms-full.txt present: full page index, descriptions of all major pages, conversational Q&A section
• humans.txt present: team credits, technology stack, ownership declaration

• ai-plugin.json manifest at /.well-known/ai-plugin.json

Note: OpenAI deprecated the plugin system in favor of GPT Actions. ai-plugin.json still has value for cross-AI discoverability but should not be treated as a primary signal.

Section 5

Crawlability & Indexation Control

Ensure efficient and accurate indexing.

• XML sitemap generated and maintained automatically
• Sitemap referenced in robots.txt
• Canonical URLs enforced on all pages
• No duplicate content accessible via alternate URLs
• Proper redirects for outdated or changed URLs (including all archived or retired pages)
• No indexing of staging or private areas
• Proper parameter handling to avoid duplicate content

• Image sitemap included for portfolio, case study, and other significant image content
• Video sitemap included when video content is present
• Preconnect links added for critical external origins
• X-Robots-Tag header support for non-HTML files

Section 6

Performance & Core Web Vitals

Maximize performance ranking signals and user experience.

• Largest Contentful Paint (LCP) under 2.5 seconds
• Interaction to Next Paint (INP) under 200 milliseconds
• Cumulative Layout Shift (CLS) under 0.1
• Images optimized with proper sizing and lazy loading
• Server-side rendering or static rendering where possible
• Minimal JavaScript payload — server components preferred
• Self-hosted fonts with preload enabled
• No layout shifts during page load or hydration
• Optimized caching strategy

Section 7

Content Structure & Semantic Architecture

Ensure machine-readable clarity and authority recognition.

• Single H1 per page
• Logical heading hierarchy (H1 → H2 → H3, no skipped levels)
• Semantic HTML5 sectioning elements: main, section, article, aside, nav
• Descriptive anchor text on all links
• Clear internal linking between related pages
• datePublished and dateModified metadata on all content pages

• Author attribution where applicable
• Visible last-updated indicator on service and informational pages
• Plain language service summaries that directly answer common natural language queries
• Conversational Q&A section in llms-full.txt addressing how, what, and who queries
• Version or revision indicators on key pages to signal active maintenance
• isPartOf and hasPart schema relationships defined where applicable

Section 8

Canonical Authority Control

Prevent ranking dilution and authority fragmentation.

• Single canonical domain enforced (www or non-www, consistently)
• HTTPS enforced — all HTTP redirected
• Duplicate URLs redirected to canonical
• Uppercase URLs redirected to lowercase
• Trailing slash inconsistencies resolved
• All archived or retired page URLs redirect to appropriate current pages

Section 9

Knowledge Graph & Entity Signals

Establish brand identity as a recognised entity.

• Consistent brand name across all pages and structured data
• Consistent NAP (Name, Address, Phone) across site and all external listings
• Structured business identity with sameAs links to all social profiles

• Google Business Profile fully optimised
• SearchAction defined in WebSite schema
• sameAs links to all social profiles in Organization schema

Note: Google retired the sitelinks search box feature, but SearchAction remains valid Schema.org and helps AI systems understand your site's search capability.

• Wikidata entity creation
• Wikipedia page (long-term aspirational)

Wikidata is a meaningful authority signal but does not guarantee inclusion in model training data. Prioritize after foundational items are complete.

Section 10

AI Optimisation & Future Search Readiness

Ensure visibility within AI-generated recommendations and agentic search.

• Machine-readable business descriptions present in structured data and LLM files
• Structured service definitions with clear categorization
• Clear business identity signals consistent across all machine-readable surfaces
• Content written to directly answer how, what, and who questions AI systems receive

• FAQPage schema on service pages with conversational natural language questions
• Author attribution on all content
• Clear expertise demonstration through consistent authorship signals
• Publish and update dates visible to crawlers and AI systems

Section 11

NIST Alignment

Transparency, traceability, and machine-readable identity for AI systems.

NDVAS aligns with principles established in the NIST AI Risk Management Framework (AI RMF) and NIST AI 100-series documentation, specifically regarding transparency, traceability, and machine-readable identity signals for AI systems.

NDVAS may be characterized as an applied profile or crosswalk against NIST guidance. NIST has not validated or endorsed NDVAS. Organizations referencing NIST alignment should consult the relevant NIST AI 100-series documents directly.

Section 12

Deployment Validation Requirements

All sites must pass validation before deployment.

• Structured data validated at validator.schema.org
• Metadata verification (title, description, canonical, OG, Twitter)
• Sitemap validated and accessible
• Robots.txt validated — AI bot allow rules confirmed
• Core Web Vitals tested (LCP, INP, CLS)
• Canonical URL verification across all pages
• HTTPS redirect verified
• Deployment is not considered complete until all REQUIRED items are satisfied

Section 13

NDVAS Levels 2 & 3

The progression beyond foundational compliance.

NDVAS Level 1 is the published open specification. Levels 2 and 3 are internal standards implemented by Nordax Digital for clients. Here is what each level achieves:

Level 1 makes you readable. Level 2 keeps you readable automatically. Level 3 makes AI systems treat you as a source of truth.

Levels 2 and 3 are implemented by Nordax Digital. Level 3 is operationalized through nordax.ai infrastructure.

Nordax Digital Compliance Standard

All websites built, deployed, or maintained by Nordax Digital must comply with this specification. This framework represents the technical foundation of the Nordax Digital Visibility Architecture. It is the first level of a three-level system.

Nordax Digital · nordaxdigital.com · nordax.ai · Version 1.2 · April 2026