llms.txt · y8y

What is llms.txt?

llms.txt is a proposed standard for a plain text file that web publishers place at the root of their website. It is read by AI language models and search systems to understand the website’s purpose, content and how to cite it accurately.

The name stands for “large language models text”: a file designed specifically for the AI systems (large language models) that power ChatGPT, Perplexity, Claude and similar services.

What goes in a hotel’s llms.txt file?

A hotel llms.txt file needs five things: property name and category, a short prose description, key attributes, proximity information, and links to the most useful pages on the site.

Property name and category (one line)
2-4 sentence prose description (what it is, where it is, who it’s for, what makes it distinctive)
Key attributes as plain text (amenities, room count, price range, policies)
Proximity information (distances to landmarks, transport links)
Links to the most useful pages on the site for AI systems to read

How does llms.txt work?

AI search systems that browse the web, Perplexity and ChatGPT with browsing, read the llms.txt file when they reach a hotel website, using it as a clean, authoritative summary of the property to draw on when generating travel recommendations.

Unlike schema.org markup (which is structured code), llms.txt is written in plain prose that AI systems can use directly in their answers.

How do I implement llms.txt?

Create a plain .txt file. Write the content as described. Upload it to your web host at the root directory so it’s accessible at yourdomain.com/llms.txt. Takes 30-60 minutes.