Free tool · 21 checks
Is your website ready for AI agents? Paste a URL. In about 10 seconds you'll see what could stop ChatGPT, Claude and other AI agents from reaching and reading your site, plus a plain-English fix for every gap.
My own site, scored
The basics
AI agent readiness is how easily an AI agent can reach your website, read it and act on it without a person clicking around for it. The agent might be ChatGPT fetching your page to answer a question, Claude researching vendors for someone, or a coding agent like Claude Code reading your docs.
For decades, websites were built for two visitors: people in browsers and search engine crawlers. Agents are a third visitor, and they read your site differently.
Here's a common example. A site built as a single-page app looks perfect in your browser. But the HTML it actually sends is an empty shell, and the text only appears after JavaScript runs. An agent that doesn't run JavaScript gets a blank page and moves on to a competitor.
So the question this checker answers is simple. When an agent shows up, does your site help it or fight it?
The opportunity
Cloudflare put numbers on this when it launched its own scanner in April 2026. Its Radar team started from the 200,000 most visited domains on the internet, filtered out redirects, ad servers and tunneling services, and found:
That gap is the opportunity. Some fixes take five minutes. Others depend on how your site is built. Either way, Cloudflare's numbers show the newer agent standards are still rare, so most sites haven't adopted them yet.
Markdown is a good example of why this matters. Agents pay for every token they read (a token is a small chunk of text). Cloudflare measured up to an 80% token reduction in some cases when an agent gets markdown instead of HTML. Fewer tokens means an agent can read more of your site before it runs out of room.
Under the hood
The checker runs 21 checks in five groups. It follows the same open standards as Cloudflare's isitagentready.com, and adds three practical checks for the problems I see most on real business sites: text hidden behind JavaScript, missing structured data, and firewalls that quietly turn AI agents away. It leaves out Cloudflare's commerce checks (x402, UCP and ACP), which don't count toward Cloudflare's score either.
Can an agent find your pages and the files that describe your site?
Can an agent read your pages without a browser, without wasting tokens?
Do you tell AI bots what they may do, and does your firewall agree?
Can an agent take an action on your site instead of just reading it?
Only scored if you run an API, an app with a login, or an MCP server.
Scoring
Each check carries a weight based on how much it affects a real agent. Text readable without JavaScript (weight 14) and AI bot rules (12) count the most, because a failure there can lock agents out completely. robots.txt (10) and your sitemap (8) come next, since they tell every bot what exists and what's allowed. Draft standards like the AI catalog and DNS-AID count the least. A warning earns half credit, and the Web Bot Auth check is shown for context but never scored.
Your report shows one score out of 100. By default it grades 13 general website checks and skips the API, login and MCP group. Most of the 13 are basics every site needs. Four are low-weight extras (llms.txt, Agent Skills, WebMCP and the AI catalog), worth 11 of the 87 weight points combined.
Default score weights
Here's why the API group is left out. On September 14, 2026, Cloudflare's scanner gave ryandoser.com a 20 out of 100. After a day of fixes with Claude Code, it reached 60 and Level 4, "Agent-Integrated." Every check still failing was for an API, a login system, an MCP server or an AI agent, none of which a blog runs. Publishing fake versions of those files to chase a number would only feed agents bad information.
If you do run an API, an app or an MCP server, tick the box in your report. The score then grades all 20 scored checks.
For the record, ryandoser.com scores 100 on the default score here and 77 with the API group counted. The gap between that 77 and Cloudflare's 60 comes from weighting. This checker puts more weight on the basics every site needs, and it adds three checks Cloudflare doesn't run, which the site passes.
Fix order
If you run a normal business, content or ecommerce site, fix things in this order. Each fix in your report comes with a prompt you can paste into Claude Code or any coding agent.
Many AI crawlers fetch the raw HTML and don't run scripts. Sites built as single-page apps (plain React or Vue, some site builders) can look empty to them. Server-side rendering or static generation fixes it.
Biggest impactCheck robots.txt and your firewall. Security plugins and CDN bot settings can block AI agents even when robots.txt welcomes them.
Biggest impactschema.org markup labels who you are, what you sell and who wrote what, so an agent doesn't have to infer it.
Biggest impactOne line. It's the cheapest way to hand an agent a list of your pages.
One more line in robots.txt with your yes or no for search, AI answers and AI training. It records your preference. Bots aren't forced to follow it.
On Cloudflare, paid plans can switch this on with a toggle. On other hosts, a small piece of middleware can serve a markdown copy of each page.
Link headers, an Agent Skills index and WebMCP tools help when you have something worth describing or an action worth offering. Skip them until the basics pass.
The AI SEO strategy guide covers the content side, and this breakdown of how Claude searches the web shows which bots fetch what.
Reality check
A perfect score won't get you recommended by ChatGPT.
Readiness covers what happens once an agent arrives. Getting named in an AI answer in the first place is a different job, and it runs on your content, your brand and what other sites say about you. Cloudflare's own launch post doesn't call its score a ranking factor, and neither should anyone selling you a fix.
Same goes for llms.txt. Google says you don't need AI text files to appear in its AI features, and there's no demonstrated citation benefit. That's why it carries one of the smallest weights here. I covered the evidence in the AI search myths post. If AI search visibility is the actual goal, start by tracking where you show up today, then look at the AI visibility tools worth paying for.
Readiness is the floor. It makes sure agents can get to the information you publish and understand it when they do.
FAQ
Run the checker above and look at three results first: Readable without JavaScript, the firewall test, and whether AI bot rules shows a search or assistant bot as blocked. If none of those are red, our scanner could reach and read the page it tested, which is a good sign for most AI agents and most of the battle for a normal business site. A yellow "no stated policy" warning on AI bot rules is worth fixing, but bots are still allowed in. Markdown, Content Signals and the agent tools are the next layer.
These are this checker's own benchmarks: 70 and up is solid for a content or business site, and 90 and up is excellent. The default score skips the API, login and MCP checks, since a blog or service business doesn't need those files. If you run an API or an app, the checkbox in your report adds them in.
Not directly. Whether ChatGPT or Google AI Overviews recommend you comes down to your content and your reputation across the web. The overlap is access. If you block AI search bots or hide your text behind JavaScript, AI search tools can't cite pages they can't read.
Cloudflare's scanner set the bar, and this checker tests the same specs. Both give you fix prompts. Two things are different here. The default score grades content sites without docking them for API and login files they don't need. And it adds three checks Cloudflare's scanner doesn't run: text without JavaScript, structured data and a firewall test.
Blocking training bots like GPTBot and ClaudeBot is a fair business call. Think hard before blocking the search bots (OAI-SearchBot, Claude-SearchBot, PerplexityBot), because they decide whether your pages can show up in AI search results. The user bots (ChatGPT-User, Claude-User, Perplexity-User) visit a page when a person asks an assistant to read it. OpenAI says robots.txt rules may not apply to those user-requested visits, so a firewall rule is the reliable way to stop them.
Free, no signup, no email. Each result is cached for 10 minutes so repeat scans and share links load fast, then it expires. The scanner only makes read-only requests to public URLs on the site you enter.
Stats on this page come from Cloudflare's April 2026 post, Introducing the Agent Readiness score. Cloudflare updates its Radar adoption data weekly, so current numbers may differ.