Agent-Readiness Auditor

See how well AI agents can read and operate any web page, with a score and specific fixes.

AI agents now browse the web for people: they compare products, fill in forms, book appointments and check out. Most websites were never designed for them. Agent-Readiness Auditor shows you, in seconds, how well agents can understand and use any page, and exactly what to fix. Open any page, click the icon, and press Run audit. You get a score from 0 to 100, a breakdown across seven areas, and a list of the specific elements that trip agents up. Each issue explains why it matters and how to fix it, and Show on page scrolls to the element and highlights it. What it checks: • Real controls: clickable divs and spans that aren't real buttons or links, links that go nowhere • Labels and names: icon-only buttons with no name, vague names like "Learn more", images without alt text • Forms: unlabeled or placeholder-only fields, missing input types and autocomplete hints, required fields that aren't marked • Interaction traps: cookie banners and overlays covering the page, CAPTCHAs and bot challenges, content in cross-origin frames, tiny click targets • Stable targeting: controls that can only be found through generated class names or auto-generated IDs • Page structure: title, language, landmarks, heading outline, page size • Machine-readable signals: meta description, Open Graph, structured data, llms.txt, and robots.txt rules for AI agents Built for front-end developers, QA and accessibility teams, SEO specialists and agencies preparing sites for agentic browsing. Many fixes also improve accessibility for people using screen readers and keyboards. Export any report as Markdown or JSON to share with your team or attach to a ticket. Private by design: audits run entirely in your browser. No account, no tracking, and nothing about the page is sent anywhere.