Methodology

What we check

AI agents are starting to browse, shop, and book on behalf of users. Your website either works for them or it doesn’t. We measure which.

The framework

We score four things: can an agent find your site, navigate it, read the content, and complete a task. Each pillar measures how well your site supports that step without relying on visual cues.

1. Discoverability

30 points

Before an agent can do anything on your site, it needs to understand what your site is and what it offers. We check whether your site provides the right signals — a summary file, structured metadata, a sitemap — so agents can orient themselves instantly.

1.1Site summary file6 pts

A text file that tells agents what your site does and how to use it.

1.3Structured data6 pts

Machine-readable markup describing your content as products, articles, or events.

1.6Sitemap6 pts

A complete map of your pages so agents don’t have to discover them by clicking.

+ 3 more criteria

2. Navigability

30 points

Once an agent knows what your site offers, it needs to move through it. We measure whether your structure is clear enough for an agent to find content, follow links, and reach the right page without getting lost.

2.1Semantic HTML6 pts

Proper landmarks and heading levels that give agents a structural outline.

2.3Link quality6 pts

Links that describe the destination, not “click here” or “learn more”.

2.4URL structure4 pts

Clean, predictable URLs agents can interpret at a glance.

+ 4 more criteria

3. Content Accessibility

20 points

Agents process code, not visual design. We check whether your content is available in clean, structured formats agents can read efficiently — without wading through scripts, ads, and interface chrome.

HTMLMD
3.2Markdown availability5 pts

Page content available in clean markdown, not just raw HTML.

3.5Gated content handling5 pts

Paywalls and logins communicated clearly, not shown as blank walls.

+ 3 more criteria

4. Task Clarity

20 points

The most advanced layer. We check whether your site helps agents actually do things — add to cart, book an appointment, complete a form — not just read content. Most sites rely on visual cues humans figure out. Agents need explicit guidance.

Add to cart
4.1Task manifest4 pts

A published list of actions agents can perform on your site.

4.2Action descriptions4 pts

Buttons that describe outcomes, not just “Submit” or “Go”.

+ 3 more criteria
Scoring

What your score means

Drag the slider to explore the five bands of agent readiness.

0/ 100
Agent-Fragile40–59

Agents can partially interact but key flows break. The foundations are there — consistency and clarity are not.

Test your site

Generate your agent readiness report →