AI Readiness Audit

What is the AI Readiness Audit?

The AI Readiness Audit is a free tool that fetches one page, along with its site's robots.txt, llms.txt and sitemap, and scores it from 0 to 100 on thirteen weighted checks. They cover AI crawler access, structured data, meta tags and answer-ready formatting. The result shows each check's score, the fixes worth the most points, and code to copy.

How to use the AI Readiness Audit

1. Enter the URL of the page you want audited, such as https://example.com/pricing. The audit reads that exact page, so pick one that matters.
2. Press Check AI readiness. The tool fetches the page's HTML, then /robots.txt, /llms.txt and the site's sitemap from a standard path such as /sitemap.xml.
3. Read the overall score out of 100, then open the thirteen check cards to see each score and the measurements behind it.
4. Work down Recommendations and the Critical Issues panel, then copy the snippets you need from the Code Implementation Center.

What does the AI readiness score measure?

It measures whether an answer engine can reach a page, read it and quote it. Thirteen checks each score 0 to 100 and carry a weight: structured data 15, content access 12, answer readiness 12, then meta tags, text quality, AI bot access and content semantics at 10 each, robots.txt and schema markup at 8, and page speed, heading structure, sitemap and llms.txt at 5.

The weights add up to 115, so the overall score is the weighted total divided by 1.15, which keeps it on a 0 to 100 scale. A check's weight is what fixing it is worth: a weak structured data score costs three times what a weak sitemap score does.

Can AI crawlers read JavaScript?

Mostly not. GPTBot, ClaudeBot and PerplexityBot fetch a page's HTML and do not run its JavaScript, so a React or Vue app that ships an empty shell and fills it in the browser looks empty to them. Googlebot does render JavaScript, which is how a page can rank in Google while AI crawlers see almost nothing.

This audit reads the served HTML the same way AI crawlers do. When that HTML carries under 500 characters of text, the Critical Issues panel says so first, because every content check is grading an empty page. Server-rendering the copy is the fix.

What makes a page answer-ready?

Content an engine can quote without rewriting it. The answer readiness check counts question-style headings (ones with a question mark, or that start with how, what, why, when, where or who), paragraphs of 20 to 200 characters, lists, tables, definition lists, step-by-step wording and comparisons such as "vs" or "difference between". It also looks for a TL;DR, key takeaways, quick answer or summary line near the top.

A 2,000-word essay with three headings and no lists reads fine to a person and gives an extractor nowhere to cut. Nothing on it is wrong. Nothing on it can be quoted on its own.

Which AI readiness fixes should I do first?

The ones at the top of Basic Improvements. That list takes every check scoring under 90, ranks its issues by the check's weight times the points it is short, and keeps the top ten, so a weak 15-point check outranks a failing 5-point one.

The Critical Issues panel lists checks scoring under 40, and Growth Opportunities lists those from 40 to 79. Clear the critical ones first. The Code Implementation Center only carries snippets for what this page is missing, such as Open Graph tags, FAQ schema or an llms.txt file.