AX Audit — AI Agent Readiness Scanner

AX Audit scans websites for AI agent readiness across 6 categories aligned to the open AgentReady AR-* standard, including AR-PROV (Agent Provisioning Readiness) — a category AINative authored and proposed upstream.

Live at: ainative.ax

Quick Start

Scan a URL (no auth required)

curl -X POST https://ainative-browser-builder.up.railway.app/api/v1/public/ax-audit \
  -H "Content-Type: application/json" \
  -d '{"url": "https://your-site.com"}'

Response:

{
  "url": "https://your-site.com",
  "score": 61,
  "max_score": 100,
  "grade": "C",
  "checks": [
    {
      "name": "MCP Discovery",
      "passed": false,
      "score": 0,
      "max_score": 12,
      "details": "No MCP discovery endpoints found"
    }
  ],
  "scanned_at": "2026-06-21T00:00:00Z"
}

6 Categories (100 points total)

Category	AR-* ID	Weight	What it checks
Discoverability	AR-DISC	12	MCP server card, .well-known/mcp.json
Agent Discoverability	AR-DISC+	15	llms.txt, agents.txt, agent.json, AGENTS.md, ai-plugin.json
Content	AR-CONT	10	JSON-LD, Schema.org, OpenGraph
Capabilities	AR-CAP	10	OpenAPI spec, /docs, /swagger, sdks.txt
Machine-Readable	-	10	robots.txt, sitemap.xml, RSS feeds
Security	-	8	security.txt (RFC 9116)
Authentication	AR-AUTH	10	OAuth 2.0, OIDC, Bearer, API keys
Error Handling	-	10	JSON error responses, proper status codes
Rate Limiting	-	8	X-RateLimit headers
SSR Detection	-	7	Server-side rendering for crawlability

Scoring Bands

Score	Grade	Band
98-100	A++	Good
95-97	A+	Good
90-94	A	Good
80-89	B	Good
70-79	C	Needs Work
50-69	D	Needs Work
0-49	F	Poor

API Endpoints

Quick Scan

POST /api/v1/public/ax-audit

No authentication. Returns score + 10 check results in ~60 seconds.

Full Audit (with simulation + benchmark)

POST /api/v1/public/ax-audit/full
Body: {"url": "https://example.com", "industry": "saas"}

Returns stored report with agent simulation transcript and industry percentile ranking.

Retrieve Report

GET /api/v1/public/ax-audit/reports/{report_id}

Industry Benchmark

GET /api/v1/public/ax-audit/benchmark
GET /api/v1/public/ax-audit/benchmark/{industry}

Industries: saas, ecommerce, fintech, healthcare, media, education, enterprise, agency.

Multi-Page Crawl

POST /api/v1/public/ax-audit/crawl
Body: {"url": "https://example.com"}

Crawls up to 500 pages (Pro tier) with per-page scores.

Checkout (Stripe)

POST /api/v1/public/ax-audit/checkout
Body: {"tier": "pro", "billing": "monthly", "site_url": "https://ainative.ax"}

Creates a Stripe checkout session. Tiers: pro, business, agency.

Pricing

Tier	Price	What you get
Free	$0	1 scan/day, full report, fix prompts
Remediation	$2,500 one-time	We fix your Discover→Authenticate gaps
Monitoring	$99/mo	Continuous re-audits, alerts, percentile tracking
Provisioning	Custom	AR-PROV conformance (the rung only AINative provides)

AR-PROV — The Category We Authored

AR-PROV (Agent Provisioning Readiness) defines what it means for a surface to be provisioning-ready: an AI agent can create and configure a new account, accept machine-readable terms, and receive scoped credentials — without a human completing a form.

• Spec: spec/categories/ar-prov.md
• Upstream PR: agentready-org/standard#2
• Discovery: /.well-known/provisioning profile + JSON Schema
• 3 conformance levels: L1 Declared, L2 Programmatic, L3 Autonomous

How to Improve Your Score

Quick wins (get to 60+)

1. Add /llms.txt — summarize your site for AI crawlers
2. Add /robots.txt — allow GPTBot, Claude-Web, PerplexityBot
3. Add /sitemap.xml — list all important pages
4. Add JSON-LD structured data to every page
5. Add /.well-known/security.txt with Contact and Expires fields

Intermediate (get to 80+)

6. Add /.well-known/mcp.json — MCP server card
7. Add /openapi.json — OpenAPI 3.1 spec
8. Add /agent.json — agent capability card
9. Add X-RateLimit-* headers
10. Ensure SSR (server-side rendering) for crawlable content

Advanced (get to 90+)

11. Implement OAuth 2.0 + PKCE with discovery endpoints
12. Return JSON error responses (not HTML) for API routes
13. Add RSS/Atom feed
14. Add Content-Signal headers
15. Support Markdown content negotiation

Integration with AINative Platform

AX Audit is part of the AINative platform. Developers on AINative can:

• Scan from Claude Code: Use the /ax-test skill to audit any site
• Monitor in dashboard: Track AX score over time
• Automate with agents: Scout agent can monitor competitor AX scores
• Use the API: Integrate scanning into CI/CD pipelines
• SEO validation: Use the /seo-validator skill to check content against Google's guidelines