# Onto — AI is reading your website wrong.
> Onto serves clean Markdown to AI agents instead of HTML noise. Same content, right format. One line of code. No changes to your site.

**Source:** /scoring
**Extracted:** 2026-03-25T16:51:29.425Z

---
Methodology

# The AIO Score

Agent Input Optimization. A single number that tells you how well AI agents can read, understand, and accurately represent your website.

Score Ranges

## What your score means

05080100

80-100

Agent-Ready

Your site is highly optimized for AI consumption. Clean semantics, low bloat, strong metadata. AI agents read you accurately and cite you confidently.

50-79

Needs Work

Some AI readability, but significant noise. Framework bloat or missing structured data is causing partial signal loss. Hallucination risk is moderate.

0-49

AI-Hostile

Heavy React Tax, no semantic structure. AI agents are hallucinating or skipping your content entirely. This is where 80% of sites we've scanned land.

The Formula

## Three pillars of AI readability

AIO = 0.4 × React Tax + 0.35 × Semantic Richness + 0.25 × Content Negotiation

Weight: 40%

### React Tax

How much of your HTML is noise?

formula: 1 - (visible\_text\_bytes / total\_html\_bytes) × 100

Measures the ratio of useless bytes (CSS classes, JS bundles, layout divs, SVGs, tracking scripts) to actual semantic text content. A 95% React Tax means only 5% of your page is meaningful to an AI agent.

< 70% taxExcellent

70-85% taxAverage

\> 85% taxPoor

Weight: 35%

### Semantic Richness

Does your HTML tell a structured story?

formula: presence\_score(headings, lists, tables, schema, main, article)

Evaluates the presence and quality of semantic HTML elements: heading hierarchy (H1-H6), lists, tables, JSON-LD structured data, <main>/<article> landmarks, and Open Graph metadata.

Rich schema + headingsExcellent

Basic headings onlyAverage

Flat div soupPoor

Weight: 25%

### Content Negotiation

Does your server speak Markdown?

formula: supports\_accept\_markdown || serves\_distinct\_agent\_payload

Tests whether your server responds differently to Accept: text/markdown or known AI User-Agents. Sites that already serve Markdown to agents get full marks. This is the frontier most sites haven't reached yet.

Serves Markdown to agentsExcellent

Robots.txt allows crawlingPartial

No agent awarenessNone

WHAT WE SCAN

## Every scan checks

HTML size vs Semantic text

Heading Hierarchy

JSON-LD Schema

Open Graph Tags

Robots.txt Access

WAF Detection

Semantic Landmarks

Content Negotiation response

Table Structure

Link Density

Script-to-Text Ratio

Meta Descriptions

## What's your score?

Enter any URL and get your full AIO breakdown in seconds.

[Scan Now](/)

---
## Structured Data (JSON-LD)
```json
{
  "@context": "https://schema.org",
  "@type": "HowTo",
  "name": "AIO Score Calculation Methodology",
  "description": "AI Optimization (AIO) Score measures how well a website is optimized for AI agents and LLM crawlers. Scored out of 100 points based on three core pillars.",
  "step": [
    {
      "@type": "HowToStep",
      "name": "React Tax (Token Efficiency)",
      "text": "Measures the ratio of useful content to total page weight. Weight: 40%. Sites with high \"React Tax\" (heavy JS/HTML/CSS noise) score lower as they consume more tokens for less information.",
      "position": 1
    },
    {
      "@type": "HowToStep",
      "name": "Semantic Richness",
      "text": "Evaluates the presence of structured metadata (JSON-LD), semantic HTML tags (<main>, <article>), and proper heading hierarchy. Weight: 35%. Essential for confident AI extraction.",
      "position": 2
    },
    {
      "@type": "HowToStep",
      "name": "Content Negotiation",
      "text": "Tests whether your server can negotiate and serve optimized Markdown payloads directly to AI agents via the Accept: text/markdown header. Weight: 25%.",
      "position": 3
    }
  ]
}
```