# 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:** /index
**Extracted:** 2026-05-21T20:21:00.768Z

---
[![](/yc.png)Backed by YC's Request for Startups](https://www.ycombinator.com/rfs#software-for-agents)Validated by[Vercel](https://vercel.com/blog/making-agent-friendly-pages-with-content-negotiation)[![](/cloudflare.png)Cloudflare](https://blog.cloudflare.com/markdown-for-agents/)

## AI is reading your website  
completely wrong.

Onto is the compatibility layer for the agent web. AI developers call our API to read any site cleanly. Site owners install our SDK to serve themselves cleanly. Agents will soon act on the web through the same engine. Same compatibility layer. Three buyer surfaces. One destination.

[Try the API free](#layers)[Scan your site](#scanner)

### One API call. From _812 KB_ of noise to _8 KB_ of signal.

POSTapi.buildonto.dev/v1/read

200 OK12 ms

x-onto-cacheHITcontent-typetext/markdownx-onto-aio96x-edgesfo3

html · before812 KB

412\-<div class\="flex items-center pt-6…"\>

413\- <script src\="/\_next/static/chunks…"

414\- <div data-radix-id\="rx-12"

415\- <span class\="sr-only"\>…

416\- <svg viewBox\="0 0 24 24"

// 11,482 more lines · 792 KB CSS + JS

Onto

markdown · after8.4 KB

1+\# Stripe — Online payments

2 

3+Accept payments in 135+ currencies

4+with a single integration.

5 

6+\## Pricing

7+\- Standard — **2.9%** + 30¢

8+\- Volume — **custom rates**

Payload**97%** smaller

Hallucination riskLow

AIO score**96**/100

Use it from your code. Use it from Claude Code, Cursor, or any MCP client.

// 39 crawlers · 19 AI providers

OpenAIGPTBot · ChatGPT-User

![](/anthropic.png)

AnthropicClaudeBot · Claude-User

![](/google.png)

GoogleGemini · Googlebot

![](/microsoft.png)

MicrosoftBingbot · Copilot

![](/meta.png)

MetaMeta-ExternalAgent

![](/xai.png)

xAIGrok

![](/perplexity.png)

PerplexityPerplexityBot

AppleApplebot · Siri

![](/amazon.png)

AmazonAmazonbot · Alexa

![](/deepseek.png)

DeepSeekDeepSeekBot

![](/tencent.png)

TencentTencentBot

![](/bytedance.png)

ByteDanceBytespider

![](/cohere.png)

Coherecohere-ai

![](/brave.png)

BraveBravebot · Leo AI

![](/exa.png)

ExaExabot

![](/you.png)

You.comYouBot

![](/duckduckgo.png)

DuckDuckGoDuckAssistBot

![](/yandex.png)

YandexYandexBot

![](/commoncrawl.png)

Common CrawlCCBot

OpenAIGPTBot · ChatGPT-User

![](/anthropic.png)

AnthropicClaudeBot · Claude-User

![](/google.png)

GoogleGemini · Googlebot

![](/microsoft.png)

MicrosoftBingbot · Copilot

![](/meta.png)

MetaMeta-ExternalAgent

![](/xai.png)

xAIGrok

![](/perplexity.png)

PerplexityPerplexityBot

AppleApplebot · Siri

![](/amazon.png)

AmazonAmazonbot · Alexa

![](/deepseek.png)

DeepSeekDeepSeekBot

![](/tencent.png)

TencentTencentBot

![](/bytedance.png)

ByteDanceBytespider

![](/cohere.png)

Coherecohere-ai

![](/brave.png)

BraveBravebot · Leo AI

![](/exa.png)

ExaExabot

![](/you.png)

You.comYouBot

![](/duckduckgo.png)

DuckDuckGoDuckAssistBot

![](/yandex.png)

YandexYandexBot

![](/commoncrawl.png)

Common CrawlCCBot

Quick AIO Scan

Drop any URL. Get your AI-readiness report in seconds.

Scan

127 sites scannedAvg score 51 / 100The web is broken for AI.

01 — The Problem

### 800 KB of HTML. 8 KB of signal.

The modern web was built for human eyes. AI agents reading the same pages parse hundreds of kilobytes of layout code to find a few kilobytes of actual content. They hallucinate. They burn token budgets. They cite the wrong information under your brand name. Onto fixes this at the protocol layer.

Without Onto

812KB

delivered to AI agents

*   792KB CSS framework classes
*   JavaScript bundles & chunks
*   Animation wrappers & divs
*   Tracking scripts & pixels
*   React hydration boilerplate

With Onto

8KB

clean Markdown payload

*   Pure semantic content
*   Structured headings & data
*   Accurate pricing & features
*   Zero hallucination risk
*   JSON-LD structured data

02 — Architecture

### Three layers. One engine.

One cleaning engine. One accuracy scoring model. One edge runtime. Read, Serve, and Act all share the same protocol-level work — and that's the moat. Anyone can ship a clean Markdown API. Only Onto is building the full compatibility layer over the existing web stack, end to end, while keeping the human experience untouched.

Live

Layer 01

#### Read

REST API + MCP server

Any AI developer turns dirty web HTML into clean, agent-ready Markdown with sub-100ms cache hits. A Firecrawl alternative with accuracy scoring, hallucination risk flagging, and MCP server support for direct integration with Claude Code, Cursor, and any MCP-compatible client.

**For:** AI developers building agents

curl

\# Firecrawl alternative
curl https://api.buildonto.dev/v1/read \\
  \-H "Authorization: Bearer $TOKEN" \\
  \-d '{"url":"https://stripe.com"}'

Alpha

Layer 02

#### Serve

Next.js SDK + edge middleware

Site owners install one line in middleware. AI crawlers get clean Markdown. Humans get the full experience, unchanged. Our own AIO score went 50 → 80. Gemini stopped hallucinating our scoring formula. Claude reads us in Markdown and explicitly confirms the MIME type.

**For:** site owners & engineering teams

\# One line install
npm install @ontosdk/next
\# That's it.

Q3 2026

Layer 03

#### Act

Browser interaction — the destination of the agent web.

When agents can read, serve, and act through one engine, the agent web becomes fully programmable. Semantic intent over brittle CSS selectors — no breakage when the UI changes. Site owners opt in. The layer no one else can ship without Read and Serve first.

**For:** autonomous agents & agent platforms

\# Coming Q3 2026
await onto.act({
  intent: "buy pro plan",
  context: "checkout flow"
})

##### Same engine. Same accuracy scoring. Same edge runtime. Same auth.

Build for agents once and never re-do it.

[Get early access](#waitlist)

MCP server · Live@ontosdk/mcp

### Already speak MCP? You're already set up.

Onto ships an official Model Context Protocol server. Drop it into Claude Code, Cursor, Cline, Zed, Continue — any MCP-compatible client. Three tools, one config block, one restart.

claude\_desktop\_config.json

{
  "mcpServers": {
    "onto": {
      "command": "npx",
      "args": \["-y", "@ontosdk/mcp"\],
      "env": {
        "ONTO\_API\_KEY": "onto\_sk\_live\_…"
      }
    }
  }
}

Works in

*   Claude Code
*   Claude Desktop
*   Cursor
*   Cline
*   Zed
*   Continue

Tools exposed

*   `read_url` — clean Markdown for any URL
*   `score_url` — AIO score & hallucination risk
*   `read_and_score` — both in one call

[Read the MCP docs](https://docs.buildonto.dev/mcp)[View on npm](https://www.npmjs.com/package/@ontosdk/mcp)

03 — Competitive Positioning

### Why not just use Firecrawl?

Firecrawl is excellent at Layer 1. But it stops there. Onto is the only company building Read, Serve, and Act with one engine — and the only one with accuracy scoring and hallucination risk flagging at every layer.

Capability

![](/onto-logo.png)Onto

![](/firecrawl.png)Firecrawl

![](/browserbase.png)Browserbase

Vercel/![](/cloudflare.png)CF

HTML → Markdown (Layer 1)

Partial

Hallucination risk scoring

AIO accuracy score

Site-side SDK (Layer 2)

Partial

Works across all hosting

Own customers only

Browser interaction (Layer 3)

Coming

MCP server

Control plane / analytics

Coming

Partial

Vercel only

04 — Industry Validation

### The infrastructure need is already proven.

Vercel, Cloudflare, Anthropic, and Y Combinator have all independently validated the thesis. Onto is how you implement it today — across any stack, any framework, any AI agent.

[

Vercel

vercel.com/blog · Feb 2026

~99% payload reduction observed.

Serving Markdown to AI agents reduces payload size by ~99%. Complex React pages drop to a few KBs of pure content.

Read blog post](https://vercel.com/blog/making-agent-friendly-pages-with-content-negotiation)[

![](/cloudflare.png)Cloudflare

blog.cloudflare.com · Feb 2026

<15ms native HTML→MD at the edge.

Markdown for Agents converts HTML at the edge — positioning AI crawlers as first-class citizens of the web.

Read blog post](https://blog.cloudflare.com/markdown-for-agents/)[

![](/anthropic.png)Anthropic

anthropic.com/news · 2023

Direct context structure → extraction accuracy.

How context is structured — instructions, document layout, the position of the question — materially changes Claude's accuracy and recall on long inputs.

Read the post](https://www.anthropic.com/news/claude-2-1-prompting)[

![](/yc.png)Y Combinator

ycombinator.com/rfs · Summer 2026

Open RFS infrastructure for the agent economy.

YC's Request for Startups explicitly calls for developer infrastructure that makes the agent economy reliable.

Read the RFS](https://www.ycombinator.com/rfs#software-for-agents)

05 — AIO Scanner

### What do AI agents see on your site?

Drop any URL. Get your full AI-readiness report in seconds. React Tax, semantic richness, hallucination risk, content negotiation status. Free. Instant. No signup.

Scan

127 sites scannedAvg score 51 / 100The web is broken for AI.

06 — Early Access

### Your AI representation is broken right now. We fix it.

Join the waitlist. Get the API key, the SDK, and the AIO Control Plane the day they ship. Founding pricing locked forever for early access customers.

Usage-based pricing for the API — pay only for what you read

Founding price for the SDK and Control Plane — $99/month, locked forever

1,000 free reads per month on the developer tier

Email address

Join

We'll only email you when we ship. Unsubscribe any time.

---
## Structured Data (JSON-LD)
```json
{
  "@context": "https://schema.org",
  "@type": "AboutPage",
  "name": "About Onto",
  "url": "https://www.buildonto.dev/",
  "description": "Onto serves clean Markdown to AI agents instead of HTML noise. It solves the React Tax by intercepting AI bot traffic and serving optimized content at the Edge."
}
```