Is Your Site Agent-Ready?
A site is agent-ready when AI agents and AI search crawlers can reach its content in plain HTML, tell the brand apart from similar names, and lift clear answers from its pages. This free check scores your site on 17 of those signals across brand, technical SEO and content, and shows which leg to fix first.
Score your AI search readiness against the agentic web standards Google previewed at I/O 2026.
Your progress saves automatically · close the tab and come back anytime.
—
out of 100 · composite Agent-Ready Score
—
—
—
Get your full 12-recommendation report
4 recommendations per leg, ordered by urgency: Quick win → 30-day → 90-day → Ongoing discipline.
No spam. One email with the report. No follow-up unless you reply.
Your 90-day path to agent-readiness.
For each leg, four recommendations ordered by urgency. Start with the Quick wins this week.
How is the Agent-Ready Score calculated?
The Agent-Ready Score is the share of 17 yes-or-no checks your site passes, scaled to 100. Five checks cover brand, seven cover technical SEO and five cover content, and “No” and “Not sure” both count as zero. A score of 80 or more is Agent-Ready, 60 to 79 is Almost ready, and under 60 is Foundation to build.
| Leg | Checks | What the checks ask |
|---|---|---|
| Brand | 5 | Whether AI answers name you, describe you consistently and tell you apart from similar names |
| Technical SEO | 7 | Organization schema and sameAs links, structured data, AI search crawler access, content without JavaScript, agent protocols and capability files |
| Content | 5 | Front-loaded claims, topic coverage, question headings, sourced data and refresh cadence |
| Score | Band |
|---|---|
| 80-100 | Agent-Ready |
| 60-79 | Almost ready |
| 0-59 | Foundation to build |
Your weakest leg is the one with the lowest share of checks passed; ties go to brand, then technical SEO, then content. The score is a self-assessment, so it is only as accurate as your answers, which is why “Not sure” counts as no.
Frequently asked questions
How do I check if my site is agent-ready?
Answer the 17 checks above and run the test printed under each one: ask ChatGPT about your brand, open your robots.txt, load your homepage with JavaScript turned off, and paste a page into Google’s Rich Results Test. Answer “Not sure” whenever you have not checked, then fix your weakest leg first.
What is a good agent-ready score?
80 or more, which this check rates Agent-Ready. Between 60 and 79, one leg is usually holding the others back; under 60, the foundations need work. Treat the score as a way to prioritize, not a benchmark: it is a self-assessment against 17 checks, not a measure of how often AI engines cite you.
Is agent-readiness the same as AI visibility?
No. Agent-readiness is whether AI agents and crawlers can reach, read and act on your site. AI visibility is whether AI answers mention and cite your brand. A site can be fully agent-ready and still rarely named, because being named also depends on third-party corroboration. The three layers are explained in SEO vs AEO vs agent-readiness.
How is this different from Lighthouse Agentic Browsing?
Lighthouse’s Agentic Browsing category, available from Chrome 150, tests one page in the browser: whether interactive elements have accessible names, whether the layout stays stable, and whether WebMCP tools are registered. Google calls it informational and unbenchmarked, so it reports passes, failures and warnings rather than a score (Chrome for Developers, June 2026). This check covers the whole program, brand and content included, and ends in a 0-100 score. Use both; see our Lighthouse Agentic Browsing guide.