AINativeainative.axAX AUDIT
← AX Audit
AX AUDIT VS. CLOUDFLARE

AX Audit vs. Cloudflare's Agent Readiness Score

Cloudflare ships agent-readiness scoring at infrastructure scale — directly in the URL Scanner and at isitagentready.com. Here's how its public scope compares to the AX Audit, and the one category it doesn't score at all.

Cloudflare's Agent Readiness Score evaluates sites across scoring categories it describes as Discoverability, Content Accessibility, Bot Access Control, Protocol Discovery, and Commerce — with Commerce called out publicly as informational only, not yet factored into the score. AX Audit scores the same discovery/content/capability/auth ground, scores commerce today, and adds AR-PROV: whether an agent can actually open and configure an account with you, not just read about you.

What Cloudflare publicly says it scores

Source: Cloudflare Changelog, 2026-05-12 — "Agent Readiness scores now available in URL Scanner."

Discoverability
robots.txt, sitemaps, AI-bot access rules
Content Accessibility
Markdown negotiation, structured content
Bot Access Control
AI-crawler allow/block rules, Web Bot Auth
Protocol Discovery
MCP server card, OpenAPI, API catalog, OAuth/OIDC discovery
Commerce
Agentic-commerce standards (x402, UCP, ACP) — informational, not scored

Category-by-category

CATEGORY
AX AUDIT
CLOUDFLARE
Discoverability (robots.txt, sitemaps, llms.txt)
Scored — AR-DISC
Scored
Machine-readable content (JSON-LD, Markdown negotiation)
Scored — AR-CONT
Scored
Capability surfaces (MCP card, OpenAPI, API catalog)
Scored — AR-CAP
Scored
Identity & delegated access (OAuth/OIDC, Web Bot Auth)
Scored — AR-AUTH
Scored
Agentic commerce (x402, ACP, UCP, MPP)
Scored — AR-COMM
Informational only — publicly stated as not yet factored into the score
Can an agent open an account with you?
Scored — AR-PROV (AINative-defined category)
Not covered
AR-PROV · THE HONEST DIFFERENCE

Can an agent actually open an account with you?

Every tool on this page — including this one — checks whether agents can find and read your site, and most now check whether they can transact with it. None of the three we compare against here define or score agent-initiated account provisioning: machine-readable terms, credential issuance, and signed-mandate consent, so an agent can sign up and get configured without a human filling out a form. AX Audit calls that AR-PROV. It's AINative's own extension to the taxonomy, not a repackaging of someone else's check.

What AX Audit checks

10+ signals across 6 AR-* categories (AR-DISC, AR-CONT, AR-CAP, AR-AUTH, AR-COMM) plus AR-PROV — the only one of the three that checks agent-provisioning readiness.

AR-DISC
Discoverability
robots.txt · A2A Agent Card · llms.txt · sitemaps
AR-CONT
Content
JSON-LD · Markdown negotiation · canonical metadata
AR-CAP
Capabilities
MCP server card · OpenAPI 3.1 · API Catalog (RFC 9727)
AR-AUTH
Identity & Access
OAuth 2.0 + PKCE · RFC 8414/9728 · OIDC · Web Bot Auth
AR-COMM
Commerce
x402 · ACP · UCP · MPP
AR-PROV · ours
Provisioning
account-create · machine-readable terms · credential issuance · signed-mandate consent
HOW WE CHECKED THIS

Verified via Cloudflare's own changelog and documentation: the five named categories, that Commerce is informational-only and not yet part of the score, and that the tool is free. We could not find a published exact count of individual checks within each category on Cloudflare's public pages, so we describe scope qualitatively rather than stating an unverified number.

See where your site actually stands

Free scan, about 60 seconds. Six categories plus AR-PROV — the one none of these tools check yet.

Run a free AX Audit →