Agent View

The site as an AI agent gets it, beside the site as a person sees it. What it can read, what it has to cut away, and what it never sees.

surfaced.so100 pages read on October 24, 2026
Worst first. 100 pages in all.
Full screen
Reset Your Password — Surfaced AI
surfaced.so/forgot-password

Image with no wordsSurfaced AI

Reset password

Forgot your password?

Enter your email and we'll send you a reset link.

© 2026 Surfaced AI

Locked out happens.

A reset link takes seconds — your saved audits and score history will be right where you left them.

AI engines we test against

The rest of the page is in full screen.

Drawn from the HTML mttr fetched, not a screenshot.
page.md70 words, about 159 tokens
[Surfaced AI](https://surfaced.so/)

Reset password

# Forgot your password?

Enter your email and we'll send you a reset link.

© 2026 Surfaced AI

## Locked out happens.

A reset link takes seconds — your saved audits and score history will be right where you left them.

AI engines we test against

![ChatGPT logo](https://surfaced.so/logos/openai.svg)![Claude logo](https://surfaced.so/logos/claude.svg)![Gemini logo](https://surfaced.so/logos/gemini.svg)![Perplexity logo](https://surfaced.so/logos/perplexity.svg)

- Track your AI Visibility Score over time
- Re-scan any site in two minutes
- Generate the full six-asset Full Pack

No <main>: Wrap the page’s own content in <main>, so a reader knows where it starts and stops, and Clean can be measured.

The figures

100 pages, set against the 168,113 sites with figures and 649 in software. Change is since the read on October 24, 2026.

See its rivals
Readablehigher is better
100%100 of 100 pages
Same as the read before
All 168,113 sitesBetter than 47%
649 software sitesBetter than 47%

Pages whose own text is in the HTML the server sends, enough of it for their kind.

Cleanhigher is better
65%middle page
Same as the read before
All 168,113 sitesBetter than 42%
649 software sitesBetter than 31%

Of the words before script, the share that are the page and not menus, footer and forms. 2 pages are not measured, with no <main> or no words.

Costshown, never scored
53,710tokens
About 537 a page

To read every page once. A long page that answers well is not worse for it. Tokens are estimated at about 4 characters each.

Readable and Clean over 4 reads

ReadableClean, middle page

Readable from 100% to 100% and Clean from 65% to 65%, between October 23, 2026 and October 24, 2026.

How its pages spread on Clean

Its middle, 65%

98 pages measured. Not measured: 2 with no <main> or no words.

Against its rivals

The sites nearest it by what they do, with figures. Close means similar work, not better.

Readinghigher is betterCostnot scored
SiteReadableCleanTokens a page
surfaced.soThis site, 100 pages100%65%537
logicballs.com100 pages100%92%4,616
rankdots.com100 pages100%99%5,814

First of 3 on Readable, third on Clean.

AI crawler access

What robots.txt says to each, as read on October 24, 2026. 8 of 8 may read the site.

Open robots.txt
Crawlerrobots.txt saysUsed forThe line that decides
GPTBotOpenAIAllowedTraining modelsUser-agent: GPTBot / Allow: /
OAI-SearchBotOpenAIAllowedChatGPT searchUser-agent: OAI-SearchBot / Allow: /
ChatGPT-UserOpenAINot namedA page a person asks aboutFalls to User-agent: *
ClaudeBotAnthropicAllowedTraining modelsUser-agent: ClaudeBot / Allow: /
Claude-SearchBotAnthropicNot namedClaude’s searchFalls to User-agent: *
PerplexityBotPerplexityAllowedPerplexity searchUser-agent: PerplexityBot / Allow: /
Google-ExtendedGoogleAllowedTraining GeminiUser-agent: Google-Extended / Allow: /
bingbotMicrosoftNot namedBing and CopilotFalls to User-agent: *

Every AI crawler mttr checks may read it. Not named means the rules for every crawler apply.

Every page

Worst first. Each opens in Agent View at the top.

2 pages of 100 lose something. Pick a loss to see what to do about it.Open the action plan

What an agent getshigher is betterSignalsCostnot scored
PageWordsCleanReadableMain headingTokensWhat it loses
/forgot-password70No <main>YesYes159No <main>
/reset-password60No <main>YesYes148No <main>
/audit2625%YesYes38Nothing
/tools11245%YesYes228Nothing
/tools/llms-txt-generator18460%YesYes302Nothing
/ai-visibility-for-builder-tools20960%YesYes379Nothing
/blog/llms-txt-guide19760%YesYes336Nothing
/help/how-often-to-audit20761%YesYes392Nothing
/help/pages-that-earn-citations21361%YesYes424Nothing
/blog/faq-schema-for-founders20461%YesYes349Nothing
/help/build-entity-profile21461%YesYes401Nothing
/help/site-speed-ai-visibility21461%YesYes428Nothing
/help/llms-txt-vs-robots-txt21862%YesYes380Nothing
/help/what-makes-content-citable22162%YesYes438Nothing
/compare/surfaced-vs-ahrefs21962%YesYes449Nothing
/help/how-the-score-is-calculated21663%YesYes411Nothing
/help/backlinks-for-ai-visibility22363%YesYes430Nothing
/help/why-my-score-changed22663%YesYes425Nothing
/ai-visibility-for-agencies22663%YesYes392Nothing
/help/ai-confuses-my-brand23663%YesYes407Nothing
/compare/surfaced-vs-peec22963%YesYes451Nothing
/compare21763%YesYes432Nothing
/help/wikipedia-and-wikidata22963%YesYes404Nothing
/help/how-ai-crawlers-discover-sites23263%YesYes450Nothing
/help/do-i-need-llms-txt23063%YesYes389Nothing
The worst 25 of 100 pages.Show all 100

Show where it stands

A card, a badge and a line, each with the figures.

surfaced.so in Agent View: 100% readable
1200 by 630, for linksDownload

A badge for the site

Agent View: 100% readable
<a href="https://mttr.si/agent-view/surfaced.so"><img src="https://mttr.si/agent-view/surfaced.so/badge.svg" alt="Agent View: 100% readable, 65% clean" height="20"></a>
Drawn again after every read

The figures in one line

surfaced.so in Agent View: 100% of pages readable, better than 47% of software sites; 65% clean; 53,710 tokens to read all 100 pages. mttr.si/agent-view/surfaced.so

What to do next

Fix the pages, let an agent do it, and read the site again to see the figures move.

Fix what agents miss

2pages lose something
  • Wrap 2 pages in <main>Clean measured
Open the action plan

Connect your agent

Claude Code, Codex or Cursor reads the site’s fixes over MCP, fixes a page in your code, has mttr read it again, and checks the fix landed.

https://mttr.si/mcp
Set up an agent

Claim the site

Prove it is yours with a DNS record or a file. Your agent can then reach its reads and jobs, and set each change to wait for your approval.

Claim the site

Drawn from the HTML mttr fetched on its read of October 24, 2026, not a screenshot, and only from the 100 pages read, where the read stopped at its page limit. An assistant’s own reader may differ. Tokens are estimated at about 4 characters each. A site an agent reads well is not a site an assistant cites.

Agent Viewsurfaced.so Close Esc
Reset Your Password — Surfaced AI
surfaced.so/forgot-password

1 of 5 images has no words, so an agent gets nothing of it.

Image with no wordsSurfaced AI

Reset password

Forgot your password?

Enter your email and we'll send you a reset link.

© 2026 Surfaced AI

Locked out happens.

A reset link takes seconds — your saved audits and score history will be right where you left them.

AI engines we test against

Image: ChatGPT logo

Image: Claude logo

Image: Gemini logo

Image: Perplexity logo

  • Track your AI Visibility Score over time
  • Re-scan any site in two minutes
  • Generate the full six-asset Full Pack
Drawn from the HTML mttr fetched, not a screenshot.
page.md70 words, about 159 tokens
[Surfaced AI](https://surfaced.so/)

Reset password

# Forgot your password?

Enter your email and we'll send you a reset link.

© 2026 Surfaced AI

## Locked out happens.

A reset link takes seconds — your saved audits and score history will be right where you left them.

AI engines we test against

![ChatGPT logo](https://surfaced.so/logos/openai.svg)![Claude logo](https://surfaced.so/logos/claude.svg)![Gemini logo](https://surfaced.so/logos/gemini.svg)![Perplexity logo](https://surfaced.so/logos/perplexity.svg)

- Track your AI Visibility Score over time
- Re-scan any site in two minutes
- Generate the full six-asset Full Pack
Drawn from the HTML mttr fetched, not a screenshot. An assistant’s own reader may differ.Tokens are estimated at about 4 characters each.