Cloudflare Agent Readiness is a component of the 📝Cloudflare AEO Suite that scores whether AI agents can discover, fetch, and understand a website, checking the domain against the emerging standards agents rely on and returning a pass, fail, or neutral result for each with an evidence trail.
It exists because the standards arrived faster than the adoption. Cloudflare's scan of the 200,000 most-visited domains found 78% serving a 📝robots.txt written for search crawlers rather than agents, 4% declaring AI usage preferences through Content Signals, 3.9% supporting Markdown content negotiation, and fewer than fifteen sites publishing MCP Server Cards.
It is aimed at site owners, technical SEO practitioners, and the engineers who will implement the fixes — anyone whose traffic increasingly arrives as an agent rather than a person.
The number that matters here isn't your score, it's the field's. Cloudflare scanned the top 200,000 domains and almost nobody passes — which means agent readiness is currently a cheap, purely technical moat, and cheap moats do not stay cheap.
Key Capabilities
- Five-category scoring — Discoverability, content accessibility, bot access control, protocol discovery, and commerce standards, each evaluated from a live scan of the domain.
- Tiered diagnostics — Failing checks are grouped by implementation effort, from quick wins through technical groundwork, advanced agent integration, and agentic commerce.
- Copy Agent Prompt — Every failing check emits a ready-made prompt describing the required code change, so a coding agent can implement the fix without translation.
- Internet-wide dataset — The same checks run weekly against the 200,000 most-visited domains on 📝Cloudflare Radar, filterable by domain category.
- URL Scanner API — An agentReadiness parameter on Cloudflare's existing URL Scanner returns the full check set programmatically for any URL you point it at.
Getting Started
- Run a free scan at isitagentready.com — no account needed, any public domain.
- In the Cloudflare dashboard, open AEO and select Agent Readiness for a zone.
- Work the failing checks in tier order, starting with the quick wins.
- Paste a failing check's Copy Agent Prompt into a coding agent to implement it.
- Read your score against the field using the weekly Radar dataset.
Related
- 📝LLMs.txt — the plain-text reading list convention the content checks look for.
- 📝Model Context Protocol (MCP) — the standard behind the MCP Server Card and capability checks.