What Is llms.txt? The AI File for Hotel Websites Explained
An llms.txt file summarises a website for AI assistants. What it can do, what it cannot, and whether a hotel needs one.
Key points
- llms.txt is a simple text file in the root directory of a website. It summarises what the site is about and links to the most important pages.
- It was proposed by the researcher Jeremy Howard in September 2024. It is an open proposal, not a binding standard.
- Google does not use it for its AI Overviews and says no new files are needed for them.
- For a hotel, it is a small, optional addition. Clear content on the website itself matters more.
For search engines, there has been robots.txt for decades. Since 2024, some have been proposing a second file, this time for AI assistants: llms.txt. Hoteliers now come across it in web agency offers. What is behind it?
What an llms.txt file is
llms.txt is a text file located at “/llms.txt” in the root directory of a website. “LLM” stands for large language model, the language models behind AI assistants such as ChatGPT. The file is meant to help them grasp a website quickly: what kind of offer is this, and which pages matter?
It was proposed by Jeremy Howard, co-founder of Answer.AI, in September 2024. A revised second version is now available. The proposal is open for feedback, but it is not a standard adopted by any standards body.¹
How it is structured
The file is written in Markdown, a simple text format. The only required element is a heading with the website’s name. This is usually followed by a short summary, a few explanatory sentences and sections with lists of links, for example to rooms, directions or meeting rooms.¹ The proposal also suggests offering important pages as plain text versions.
For a hotel, it might look like this: the heading “Lakeside Country Hotel”, below it one sentence on the location, number of rooms and who the property suits. Then come the sections “Rooms”, “Getting here” and “Offers”, each with a link and a short description of the page. Nothing more is needed.
What it can and cannot do
llms.txt is an offer, not a control. Unlike robots.txt, it does not determine who may read a website. Whether an AI assistant fetches the file at all is up to each provider.
Google is clear on this: its AI Overviews and AI Mode require no new machine-readable files or AI text files. What matters is that pages are indexed and the basics of search engine optimisation are right.² Google representatives also said publicly in 2025 that Search does not use llms.txt.³
Does a hotel need an llms.txt file?
It is not mandatory, and it does not replace a good website. Anyone who describes their content clearly and specifically on their website has already done the most important part. The file cannot do any harm, though, and the effort is small: it can be written in an hour. Some website systems now generate it automatically. eingecheckt.de itself offers an llms.txt file. Whether it contributes to visibility in ChatGPT cannot be measured separately.
What hoteliers should look for
- Content first: Describe the homepage, rooms, location and offers clearly first. The file only points to them.
- Keep it short: Name, two or three sentences about the property, links to the most important pages. No marketing language.
- Keep it current: When pages or offers change, update the file too.
- Do not overrate it: Anyone who promises positions in ChatGPT with an llms.txt file is promising too much.
- Check: After uploading, open “/llms.txt” in the browser and make sure the links work.
More in the magazine
What AI visibility is and how to measure it is explained in the glossary entry on AI visibility; how to prepare content for AI answers is covered in the entry on GEO.
This glossary entry is based on the following sources:
-
Jeremy Howard (2024, revised 2026)
The /llms.txt file
Proposal, llmstxt.org. Purpose, file structure, text versions of pages, status as an open proposal. -
Google Search Central (2025)
AI features and your website
Documentation. Conditions for AI Overviews and AI Mode, no additional technical requirements, no new files or special structured data, SEO fundamentals still apply. -
Search Engine Land (2025)
Google says normal SEO works for ranking in AI Overviews and llms.txt won’t be used
Report. Statements by Gary Illyes (Google) at Search Central Live, July 2025.
More terms
- What Is a Hotel PMS? Property Management Systems Explained
- What Is a Channel Manager? Hotel Distribution Across Platforms Explained
- What Is a Booking Engine? Direct Booking on the Hotel Website Explained
- What Is CRM? Guest Data for Hotels Explained
- What Is a CRS? Hotel Central Reservation Systems Explained
- What Is Digital Check-in? Hotel Self Check-in Explained
- What Is a Hotel POS System? Tills and Point of Sale Explained
- What Is an RMS? Hotel Revenue Management Systems Explained
- What Is llms.txt? The AI File for Hotel Websites Explained
No term in this section matches your search.
- What Is an OTA? Online Travel Agencies for Hotels Explained
- What Is a Direct Booking? Direct Booking Share for Hotels Explained
- What Is a GDS? Global Distribution Systems for Hotels Explained
- What Is Upselling? Hotel Upgrades and Extras Explained
- What Is Online Reputation? Hotel Reviews Explained
- What Is a No-Show? Non-Arrivals and Overbooking Explained
- What Is Metasearch? Hotel Price Comparison Explained
- What Is Rate Parity? Hotel Parity Clauses Explained
- What Is a Rate Plan? Hotel Rate Structures Explained
- What Is AI Visibility? AI Visibility for Hotels Explained
- What Is GEO? Generative Engine Optimization for Hotels Explained
No term in this section matches your search.
- What Is Hotel Occupancy? Room and Bed Occupancy Explained
- What Is ADR? Average Daily Rate for Hotels Explained
- What Is RevPAR? Revenue per Available Room Explained
- What Is Revenue Management? Hotel Pricing Explained
- What Is TRevPAR? Total Revenue per Available Room Explained
- What Are Hotel Operating Models? Lease, Management and Franchise
- What Is Pick-up? Hotel Booking Pace and Lead Time Explained
- What Is GOPPAR? Profit per Available Room Explained
No term in this section matches your search.
