Reference Scan: docs.cloudflare.com

An example of an agent-ready site — used as the reference implementation for this scanner. Live results may drift as the site changes; re-run any time with the scanner.

Level 3 — Agent-Ready

Score: 8/9 · Scanned 2026-08-14 14:17 UTC

✅ Discoverability5/5

✓robots.txtPresent (200)

✓SitemapReferenced in robots.txt: https://developers.cloudflare.com/sitemap-index.xml

✓llms.txtFound at /llms.txt

✓MCP server cardFound at /.well-known/mcp/server-card.json

✓Link headersHomepage advertises Link: </api/>; rel="service-doc" (RFC 8288)

✅ Content1/1

✓Markdown negotiationReturns text/markdown when requested via Accept header

⚠️ Bot Access Control2/3

✗AI bot rulesrobots.txt has no explicit GPTBot/ClaudeBot rules — Cloudflare blocks nothing, so it never names bots

✓AI bots not blockedHomepage serves 200 to GPTBot user-agent

✓Content Signals1 Content-Signal directive found

📡 API, Auth, MCP & Skill Discovery2/9 (informational)

✓Agent SkillsIndex at /.well-known/skills/index.json

✓API CatalogRFC 9727 at /.well-known/api-catalog with correct application/linkset+json content type

✗OAuth/OIDC discoveryNo metadata at either well-known path

✗OAuth Protected ResourceNo /.well-known/oauth-protected-resource

✗Auth.md agent registrationNo /auth.md

✗Web Bot Auth directoryNo /.well-known/http-message-signatures-directory

✗AGENTS.mdNo /AGENTS.md file

✗agents.jsonNo /.well-known/agents.json

✗DNS-AIDNo records at _index._agents., _a2a._agents., or _mcp._agents. entrypoints

💰 Commerce0/4 (informational)

✗x402, MPP, UCP, ACPNone detected — expected, it's a docs site, not a commerce site

The one thing they're missing: their only scored gap is explicit AI bot rules in robots.txt. Since they don't block any crawlers, they never name them — a pass-by-default situation our check doesn't credit. That's the check that keeps them at 8/9 instead of 9/9.

View the raw scan JSON (for later reference)

Generated by the Agent-Ready Scanner · Re-scan this site live