llms.txt for Cyprus Businesses: What It Does, Why Google Ignores It, and an Example File

llms.txt is a plain-text Markdown file at a website's root that summarises the site and lists its main pages for AI tools. Jeremy Howard proposed it in September 2024, and it is not an official standard. Google Search does not use it: Google says the file will "neither harm nor help" your rankings, "as Google Search ignores them". Coding agents do read it. For a Cyprus business it is a 20-minute job that belongs after crawler access and readable pages.

This guide first ran in June 2026 with a blunt verdict: an llms.txt file is hygiene, not a lever. Since then Ahrefs has measured who requests the files, and Google's guide to its generative AI features, updated in July 2026, spells out that Search ignores them. Both confirm the verdict. This version keeps it, adds the evidence, shows a real file, and sets out the order of work for a Cyprus business.

What is an llms.txt file?

An llms.txt file is a Markdown file at yourdomain.com/llms.txt that gives AI tools a short, clean account of a website: what it is and which pages matter. Jeremy Howard published the proposal at llmstxt.org on 3 September 2024, "to provide information to help agents use a website". A second version followed on 10 August 2026, revised after two years of adoption.

The format is deliberately small:

• An H1 with the name of the site. This is the only required part.
• A blockquote with a short summary.
• Optional paragraphs of detail.
• Sections under H2 headings, each a list of links with a one-line note.
• By convention, a section called "Optional" for links a tool can skip when it is short of space.

Version 2 adds two things: a way for any page to point to a clean Markdown copy of itself, and a rule that a file placed under a subfolder covers the pages beneath it. Neither changes what a small business needs to do. It remains a proposal: no standards body such as the W3C or the IETF has adopted it.

Does Google use llms.txt?

No. Google Search ignores llms.txt, and Google says so in writing. The Google guide to optimising for generative AI features, last updated on 10 July 2026, has a section headed "Mythbusting generative AI search: what you don't need to do". One item in it covers "LLMS.txt files and other 'special' markup":

"You don't need to create new machine readable files, AI text files, markup, or Markdown to appear in Google Search (including its generative AI capabilities), as Google Search itself doesn't use them."

And, on llms.txt by name: "It's completely fine if you decide to create and maintain LLMS.txt files (or other similar files) for other services or systems that use these files. Doing so will neither harm nor help your site's visibility or rankings in Google Search, as Google Search ignores them."

That settles the question of llms.txt and SEO, as far as Google goes. The file does nothing for rankings, AI Overviews or AI Mode. If Googlebot shows up in your logs fetching it, that is not a signal either: Google notes that crawling a file does not mean it is treated in a special way. Anyone selling llms.txt as a Google factor is contradicting Google's own documentation.

Which AI systems read llms.txt?

Almost none of the answer engines, on the one large measurement we have found. Ahrefs studied 137,210 domains that received traffic in May 2026 and published the results on 15 June 2026:

• 28% of the domains served a valid llms.txt file.
• 97% of those files received no requests at all during the month.
• Of the requests that did arrive, 19.5% came from AI bots, and 1.1% from AI retrieval bots, the kind that fetch pages to answer a person's question. OpenAI's search crawler, OAI-SearchBot, made 0.74%.
• "Zero requests came from AI bots for llms.txt files that don't exist. They never go looking."

That last line is the one to weigh when deciding whether to bother. An engine does not look for your llms.txt, and it does not miss one that is absent. Where the file is read, it is read by tools that already know to ask: the proposal's own site says that coding agents "use them reliably".

So the strong case for the file is narrow. A software company whose documentation developers load into coding tools should have one. For a hotel, a law firm or an accounting practice in Limassol, the file is harmless and cheap, and nothing more.

How is llms.txt different from robots.txt and a sitemap?

robots.txt controls access, a sitemap lists your URLs, and llms.txt describes the site. Google Search uses the first two and ignores the third.

The one hard dependency runs through robots.txt. If it blocks a crawler, nothing on the site reaches that crawler, llms.txt included. Open yourdomain.com/robots.txt and look for either of these:

User-agent: GPTBot
Disallow: /

User-agent: *
Disallow: /

The first turns away OpenAI's training crawler. The second turns away every crawler that obeys the file. A rule written years ago against scrapers is the usual culprit, and the cost is real: OpenAI's crawler documentation states that sites opted out of OAI-SearchBot "will not be shown in ChatGPT search answers". If a wider rule is in place, allow the search crawlers by name:

User-agent: OAI-SearchBot
Allow: /

User-agent: PerplexityBot
Allow: /

User-agent: Bingbot
Allow: /

Our own robots.txt names 17 crawlers and allows every one of them.

What does an llms.txt file look like? A real example

Here is the start of ours, served at axenorconsulting.com/llms.txt, with the lists shortened:

# Axenor Consulting

> Axenor Consulting is a digital agency based in Limassol, Cyprus, working with clients worldwide.
> It builds the systems a business runs on (websites, client portals, automations, production software)
> and runs the work that brings demand to them (SEO and AI search, social media, paid media, branding).

## Grow

- [SEO & AI Search](https://axenorconsulting.com/seo): Technical SEO, content and structured data, including GEO and AEO work for AI answer engines.
- [Social Media](https://axenorconsulting.com/social-media): Social strategy, content production, scheduling and community management, reported monthly.

## Tools

- [AI readability check](https://axenorconsulting.com/ai-readability-check): A public tool that fetches a website the way an AI crawler does and reports crawler access in robots.txt, text present before JavaScript runs, indexability, canonical, structured data, sitemap and Wikidata presence.

## Contact

- Email: info@axenorconsulting.com
- Location: Limassol, Cyprus. Clients worldwide.

Three choices make it useful. The summary is third person and factual, with no adjectives a reader would have to discount. Each link carries one line written for a reader that will not open the page. And the file is generated from the same data as the site's menu and article index, so a new service or guide appears in it the day it goes live, and every line matches a page that says the same thing.

For a Cyprus hotel the sections would be Rooms, Location and getting here, Dining, Policies and Contact. For a law firm: Practice areas, People, Guides and Contact.

How do you create an llms.txt file?

1. Write it in a plain-text editor. Notepad, TextEdit in plain-text mode or a code editor. A word processor adds formatting that breaks the file.
2. Start with an H1 and a one-paragraph summary. Third person and factual: what the business does, for whom, and where.
3. List 5 to 15 key pages under H2 headings, each in the form - [Page title](full URL): one-line description.
4. Upload it to the site root as llms.txt, next to robots.txt, and check that https://yourdomain.com/llms.txt opens as plain text with no login.
5. Keep it in step with the site. Update it when a service, a page or the address changes. A file that contradicts the site gives any tool that reads it the wrong answer. If your platform can generate it from the site's own data, as ours does, let it.

Is llms.txt worth it for a Cyprus business?

Yes, as a 20-minute job, and only after the work that decides far more. Our own audit shows the gap. In June 2026 Axenor Consulting checked the websites of 10 Cyprus corporate-services firms and four Paphos hotels. Seven of the ten firms had no proper structured data. None had a Wikidata or Wikipedia entity, and none had FAQ markup. Only one of the four hotels described itself as a hotel in structured data, and the five-star property had none. A missing llms.txt was the least of what those sites lacked.

The order of work we use:

1. Crawler access. The robots.txt check above.
2. Text in the HTML. The business's name, services and location readable before any JavaScript runs.
3. A clear identity. The same name, address and description everywhere, backed by entity signals such as a Wikidata entry.
4. Pages kept current. A quarterly refresh of the pages that matter, with honest update dates.
5. Then llms.txt.

Our AI visibility checker tests the first three from outside, the way a crawler sees the site. It does not test for llms.txt. On our work page, a Limassol tax and advisory firm got its llms.txt in the same pass as a sitemap cut by three quarters, a diagnosis of its unindexed pages and a new Wikidata entity. The file was the smallest item on the list, which is where it belongs in any SEO and AI search engagement.

Key takeaways

• llms.txt is a Markdown summary of a site for AI tools, proposed in 2024. It is not an official standard.
• Google Search ignores it. Google says the file will "neither harm nor help" visibility or rankings.
• In Ahrefs' June 2026 study, 97% of llms.txt files received no requests in a month, and AI retrieval bots made 1.1% of the requests that did arrive.
• Coding agents read it. For a hotel or a professional firm it is cheap hygiene.
• Fix crawler access, readable HTML and entity signals first. The file comes last.

Frequently asked questions

Is llms.txt good for SEO? Not for Google. Google's documentation says the file "will neither harm nor help" a site's visibility or rankings in Google Search, because Search ignores it, and that covers AI Overviews and AI Mode too. Its value, where there is any, lies with other AI tools and coding agents that choose to read it.

Do ChatGPT and Perplexity read llms.txt? Rarely, on current evidence. In Ahrefs' study of 137,210 domains, AI retrieval bots, the ones that fetch pages to answer a question, made 1.1% of the requests for llms.txt files in May 2026, and OpenAI's search crawler made 0.74%. We have found no statement from OpenAI or Perplexity that either uses the file to choose sources.

Is llms.txt an official standard? No. It is a proposal published at llmstxt.org by Jeremy Howard in September 2024 and revised as version 2 in August 2026. No standards body such as the W3C or the IETF has adopted it. Documentation platforms now generate the file automatically, and Ahrefs found one on 28% of the domains it checked.

Where does the llms.txt file go? At the root of the domain, next to robots.txt, so that it opens at yourdomain.com/llms.txt. The name is lowercase. It must load without a login and return plain text. Version 2 of the proposal also allows a file in a subfolder, covering the pages beneath it.

When should I update my llms.txt file? Whenever a service, a key page or the business's details change, and at least once a quarter when you review the site. The file should say exactly what the pages say. If your site can generate it from the same data as its menus and page lists, it stays current without anyone remembering to edit it.