# AI Readiness: 22 Checks

> Complete list of all 22 checks in the Be1st.ai AI Readiness tool.

22 checks total

> Checks marked * are context-sensitive - may be N/A depending on site type.
>
> Checks marked [BONUS] never penalize.

## AI Crawlability & Access

*A1 to A5 - structured data and AI access*

### A1 - Organization schema

Organization or LocalBusiness schema with company name, logo, and contact information.

### A2 - Product schema *

Product schema for e-shops with price, availability, and ratings.

### A3 - FAQ schema *

FAQPage schema - increases the chance of appearing in AI answers.

### A4 - Review/Rating schema *

Review or AggregateRating schema for products and services.

### A5 - AI bot policy

Policy for GPTBot, ClaudeBot, CCBot, Google-Extended, PerplexityBot in robots.txt. Evaluates blocking and consistency.

## AI Content Structure

*A6 to A13 - llms.txt, content structure and quality for AI*

### A6 - llms.txt existence

Checks for the existence of /llms.txt - a direct channel for AI crawlers.

### A7 - llms.txt content quality *

Evaluates llms.txt quality - site description, section links, .md references. N/A if A6=fail.

### A8 - llms-full.txt (bonus) * [BONUS]

Bonus check - complete content export for LLM crawlers. Never penalizes.

### A9 - BLUF (Bottom Line Up Front)

Whether the main idea is in the first 40-80 words below H1 - AI extracts from there.

### A10 - Content structure (lists, tables)

Use of lists and tables for better AI processing.

### A11 - FAQ section on page *

Presence of a FAQ section on the page - a source for AI answers.

### A12 - Definitions / glossary *

Use of <dl> or definition patterns for precise explanations of terms.

### A13 - Knowledge chunkability & content depth

Content divisibility into chunks + depth and readability of text.

## AI Trust & Entity Signals

*A14 to A16 - trustworthiness and entity signals*

### A14 - Reference / evidence signals

Numbers, statistics, citations, and external sources strengthening content credibility.

### A15 - Freshness signals *

Date signals: datePublished, dateModified, <time> tags - content freshness.

### A16 - Entity & author completeness

E-E-A-T signals: About page, contact, Person schema, privacy policy, terms of service, address.

## AI Delivery Formats & Knowledge Access

*A17 to A19 and A21: formats and access to knowledge*

### A17 - AI-friendly content formats & feeds

.md links, RSS/Atom feed, JSON API: machine-readable access to content.

### A18 - Linkability of key facts

Whether key facts are directly linkable (anchor links, IDs on headings).

### A19 - Changelog / release notes *

Presence of a changelog or release notes: a signal of regular updates.

### A21 - Markdown for AI agents *

Whether the site can serve this page as Markdown: via Accept: text/markdown or an alternate .md URL. Not llms.txt (A6/A7), this is about the page itself.

## Technical AI Signals

*A20 and A22: technical readiness*

### A20 - Semantic HTML

Use of <article>, <section>, <nav>, <aside>, <main>, <figure> - 3+ = pass.

### A22 - Dashes in text (AI writing tell)

Looks for dash characters in the title, meta description, headings, body text, attributes, JSON-LD and llms.txt. For every hit it says exactly where it sits. Informational, it does not lower the score.

---

## Scoring Methodology

5 states: pass (100pts), partial (50pts), warning (25pts), fail (0pts), n/a (excluded from average)

Weighted average - each check has a weight of 1-5 based on importance.

`Score = Σ(result × weight) / Σ(weights of applied checks) × 100`

### Classification

| Grade | Score | Label |
|---|---|---|
| A+ | 90–100 | Excellently prepared |
| A | 75–89 | Well prepared |
| B | 55–74 | Partially prepared |
| C | 35–54 | Basic readiness |
| D | 15–34 | Poorly prepared |
| F | 0–14 | Not prepared |