AI Agent Readiness Checker
Audit any page for the signals AI agents and crawlers look for: llms.txt, robots.txt rules, structured data, landmarks and form signals. Scored to 100, with fix guidance for every check.
Audit a URL
The server fetches the page, /llms.txt, /robots.txt and /sitemap.xml read-only (GET, 10 second timeout, size capped). Private and internal addresses are refused. Results only ever cover what these fetches can see.
Fetching the page and its machine-readable files…
What agents actually look at
An AI agent meeting your site for the first time does three things fast: it checks whether it is allowed in (robots.txt, per-crawler rules), it looks for a machine summary (llms.txt and sitemap.xml), and it tries to understand the page itself. That last step works best on pages built the accessible way: one clear title, a real heading outline, landmarks for header, main content, navigation and footer, and forms whose inputs have names and labels.
None of this is exotic. It is the same hygiene that helps search engines and screen readers, pointed at a new audience. The score above measures the auditable half; the other half, JavaScript behavior and real agent interactions, no fetch-based checker can grade honestly, so this one does not pretend to.
Frequently Asked Questions
What is WebMCP?
WebMCP is an emerging convention that lets web pages expose machine-usable actions, such as forms with clear names and labeled inputs, so AI agents can complete tasks on a site instead of only reading it. The signals this checker looks for, like labeled forms and semantic structure, are the building blocks that make such agent interactions possible.
What does llms.txt do?
llms.txt is a proposed markdown file placed at the root of a website, usually at /llms.txt, that gives AI models a clean summary of the site: what it is, and a structured list of its important pages. It is the machine-readable sibling of a sitemap, written for language models instead of search crawlers.
Why do AI agents need structured pages?
AI agents parse pages the way screen readers do: landmarks tell them where the navigation, main content and footer are, headings tell them the topic outline, and labeled form inputs tell them what each field expects. A page that is structured for humans with headings and labels is already halfway to being structured for agents.
What does this checker NOT verify?
The checker only inspects what a simple fetch can see: the HTML, llms.txt, robots.txt and sitemap.xml. It cannot verify JavaScript-rendered content, actual crawler behavior, WebMCP implementations, or how any specific AI agent treats your site. Scores are signals, not certifications.
Is any of my data uploaded?
The URLs you audit are fetched by the Cubbbix server to run the checks, so the audited site sees a request from Cubbbix. Nothing you enter is stored or shared beyond the audit itself, and nothing is uploaded from your device.
Check Out Other Popular Tools
JXL Resizer — Resize JPEG XL Images Online, Free
Resize JPEG XL (.jxl) images online for free: set exact pixels or scale by percent, save as JXL, PNG or JPG. Private in-browser resizing, no uploads.
IELTS Band Score Calculator
Accurately calculate your IELTS band score for Listening, Academic Reading, General Reading, and Overall band score online.
French Days & Months
Interactive reference for French days of the week and months of the year. Features audio pronunciation, IPA phonetics, gender rules, prepositions, and date sentence builder.
Was this tool helpful?
Comments
Loading comments...