UAI-1 is the current public message standard for structured AI-to-AI communication.
UAIX shield mark Public interoperability charter UAIX.org Universal Artificial Intelligence Exchange Public standards publication
Current UAI-1
  • Current version UAI-1
  • Plain-English definition Open message format for auditable AI-to-AI exchange
  • Current attribution Michael Joseph Kappel, MCP
  • Start here Get Started, Project Handoff, UAI-1, validator, roadmap, and changelog

AI-Ready Web

AI-Ready Web Volume 6: Validator and Testing

Readiness assessment categories, evidence model, automated checks, manual review, score boundaries, no-op simulation, and source tests.

  • Record UAIX-DOC-3579
  • Path /en-us/ai-ready-web/validator-testing/
  • Use Canonical public record

Document status

Public standards page Published on UAIX as part of the current public standards record
Code
UAIX-DOC-3579
Surface
AI-Ready Web
Access
Public and linkable

How to use this page

Use this page for AI-Ready Web readiness assessment categories, automated checks, manual review, scoring boundaries, and no-op simulation.

Summary

Volume 6 defines how a site should test AI-readiness without turning a checklist into a certification claim. Automated checks are useful, but manual accessibility, privacy, security, and support-claim review remain required before publication.

Validator categories

Category Automated checks Manual review
Discovery Fetch robots, sitemap, .well-known, llms advisory file, route inventory, schemas, OpenAPI. Confirm files are public-safe and do not expose private routes.
Accessibility HTML validity, headings, labels, names, focusable controls, contrast where tooling supports it. Keyboard task run, screen-reader smoke pass, hidden-content review.
APIs OpenAPI parse, method/content-type checks, Problem Details shape, idempotency markers. Auth/consent review, destructive-action review, rate-limit and abuse posture.
Capability Capability matrix exists, fallback paths exist, no-op triggers exist. Confirm lowest safe path and human review path match real operations.
Privacy and rights No obvious secrets or private hostnames; policy links exist. Data minimization, PII exposure, rights signals, retention, and jurisdiction review.
Evidence Readiness result validates and cites route evidence. Changelog, release notes, exception review, and owner approval.

Scoring principle

Scores are advisory. A site with one critical blocker is not ready for the affected profile, even if its numeric score is high. The readiness result schema therefore records blockers, warnings, skipped checks, and manual review state separately from weighted scores.

No-op simulation

Every validator run should include at least one unsupported-action scenario. The correct result is a bounded response: no hidden route guessing, no credential request, no destructive action, a human review URL, and a reason code.

UAIX source test

UAIX source now includes a focused local test for this program at scripts/test-uaix-ai-ready-web-assets.ps1. It parses all new JSON assets, checks the static discovery references, verifies requirement IDs and maturity labels, and blocks certification or runtime-authority overclaims.