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.
AboutNewsEventsOpen sourceSustainabilityWorkServicesAgentic systems
Insights
/
Article
Nearform on LinkedIn ›
Graphql. Accelerated - Matteo Collina: Tech Talk Video
Image: avatar
Image: avatar
2 Apr 2020
The rest of the page is in full screen.
[Nearform on LinkedIn](https://www.linkedin.com/company/nearform/) # Graphql. Accelerated - Matteo Collina: Tech Talk Video   2 Apr 2020 - Share - - - - ## Using GraphQL to overcome architecture bottlenecks GraphQL is a query language that is rapidly gaining wide adoption across the community. It combines type validation with a query and filtering syntax that makes it easy to get up-and-running with a powerful web API in almost no time. Features like running parallel queries or update-all become much easier because they are first-class citizens of GraphQL. Add to that a vibrant community that keeps creating excellent tooling and documentation, it’s clear why GraphQL has become so popular with developers Every abstraction has a cost, and GraphQL is no exception. The added complexity and a new schema format to parse and execute mean new performance bottlenecks. In addition to performance issues, the wrong use of GraphQL can lead to architectural bottlenecks. Instead of viewing this as a problem, I took this as a challenge. As a Technical Director at NearForm, Matteo Collina consults for some of the top brands of the world. Matteo is a member of the [Node.js Technical Steering Committee](https://github.com/nodejs/tsc) focusing on streams, diagnostics and http. He is also the author of Node.js MQTT Broker, Mosca, the fast logger [Pino](https://github.com/pinojs/pino) and of the [Fastify](https://www.fastify.io/) web framework. Recorded March 26th 2020 | Work From Home Conf Organised by James Snell - Head of NearForm Research In partnership with NearForm & GitHub. More details [here.](https://community.nearform.com/wfh-conf) Need GraphQL experts for your next project? [Contact us](https://nearform.com/contact/) today to see how we can help! [←NearForm hosts First 24-hour Live WFH Co...](https://nearform.com/insights/nearform-hosts-first-24-hour-live-wfh-conference)[3 digital transformation strategy insigh...→](https://nearform.com/insights/3-digital-transformation-strategy-insights) Insight, imagination and expertly engineered solutions to accelerate and sustain progress. [Contact](https://nearform.com/contact)
An agent gets this page whole: its own text is in the HTML the server sends.
The figures
100 pages, set against the 167,893 sites with figures.
- Readablehigher is better
- 100%100 of 100 pages
- All 167,893 sitesBetter than 47%
Pages whose own text is in the HTML the server sends, enough of it for their kind.
- Cleanhigher is better
- 95%middle page
- All 167,893 sitesBetter than 93%
Of the words before script, the share that are the page and not menus, footer and forms.
- Costshown, never scored
- 254,439tokens
- About 2,544 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.
How its pages spread on Clean
100 pages measured.
Against its rivals
The sites nearest it by what they do, with figures. Close means similar work, not better.
| Readinghigher is better | Costnot scored | ||
|---|---|---|---|
| Site | Readable | Clean | Tokens a page |
| formidable.comThis site, 100 pages | 100% | 95% | 2,544 |
| techempower.com98 pages | 99% | 95% | 3,842 |
| itpathsolutions.com100 pages | 98% | 96% | 6,200 |
| evilmartians.com100 pages | 100% | 97% | 4,378 |
First of 4 on Readable, third on Clean.
AI crawler access
What robots.txt says to each, as read on October 23, 2026. 8 of 8 may read the site.
| Crawler | robots.txt says | Used for | The line that decides |
|---|---|---|---|
| GPTBotOpenAI | Not named | Training models | Falls to User-agent: * |
| OAI-SearchBotOpenAI | Not named | ChatGPT search | Falls to User-agent: * |
| ChatGPT-UserOpenAI | Not named | A page a person asks about | Falls to User-agent: * |
| ClaudeBotAnthropic | Not named | Training models | Falls to User-agent: * |
| Claude-SearchBotAnthropic | Not named | Claude’s search | Falls to User-agent: * |
| PerplexityBotPerplexity | Not named | Perplexity search | Falls to User-agent: * |
| Google-ExtendedGoogle | Not named | Training Gemini | Falls to User-agent: * |
| bingbotMicrosoft | Not named | Bing and Copilot | Falls 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.
4 pages of 100 lose something. Pick a loss to see what to do about it.Open the action plan
Show where it stands
A card, a badge and a line, each with the figures.
A badge for the site
<a href="https://mttr.si/agent-view/formidable.com"><img src="https://mttr.si/agent-view/formidable.com/badge.svg" alt="Agent View: 100% readable, 95% clean" height="20"></a>
The figures in one line
formidable.com in Agent View: 100% of pages readable, better than 47% of sites; 95% clean; 254,439 tokens to read all 100 pages. mttr.si/agent-view/formidable.com
What to do next
Fix the pages, let an agent do it, and read the site again to see the figures move.
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/mcpSet 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 siteDrawn from the HTML mttr fetched on its read of October 23, 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.