llms.txt: the file that helps AIs understand your website (technical guide)

llms.txt is a plain-text file placed at the root of your website that lists, in simple language, the most important pages of your clinic. It doesn't guarantee citations in AIs, but it works as a welcome map for language models: it shows, in an organized way, the best your site has to offer to be read and summarized.

At Bincoo, when reviewing the structure of thousands of clinic websites, we saw that the best candidates for citation have two things in common: clear pages and an architecture that any reader, human or machine, understands in seconds. llms.txt attacks exactly the second part.

More than 10,000 companies already use Bincoo to be recommended by ChatGPT, Gemini, and other platforms.

How llms.txt works

The file lives at a simple address, the path followed by convention, and is written in Markdown, the same format as this guide. The standard structure has:

1. A title and a short summary of who the clinic is;
2. A section with the most important pages, called mandatory;
3. A section with complementary content, called optional;
4. Links with one-line descriptions, explaining what the reader will find on each page.

The idea is that a language model, upon receiving this file, understands in seconds what the site offers and where to go deeper.

What to put in your clinic's llms.txt

Think of the file as the menu of what you want the AI to know about you:

• Page for each service: implants, orthodontics, aesthetics, checkups;
• Team page, with credentials and professional registrations;
• Articles that answer frequently asked patient questions;
• Contact information and address, identical to Google's;
• Reviews and mentions on portals, if there's a dedicated page.

llms.txt, sitemap, and robots.txt: who does what

File	Format	For whom	Function
sitemap.xml	XML	Search engine bots	List of URLs for crawling
robots.txt	Text	Search engine bots	Access and blocking rules
llms.txt	Markdown	Language models	Readable map of the best content
Schema	Code on the page	Search engines and AIs	Structured data about the clinic

None replaces the others. llms.txt comes in as an extra layer of clarity, especially useful for agents and tools that read websites on behalf of users.

Step by step: publish yours in 4 stages

1. List the pages that deserve prominence

Select 10 to 20 URLs: services, team, main articles, and contact. Quality matters more than quantity.

2. Write short, specific descriptions

One line per page, saying what the reader finds there. Avoid marketing adjectives; prefer facts.

3. Publish at the domain root

The file must be publicly accessible, at the standard path followed by the convention.

4. Keep it updated

Every important new page goes into the file. An outdated llms.txt passes wrong information to whoever reads it.

Common mistakes

• Filling the file with links without descriptions;
• Listing generic pages and forgetting the service ones;
• Describing with promises the site doesn't keep;
• Publishing and never touching the file again;
• Treating llms.txt as a substitute for good content.

The part the file doesn't solve

No technical file turns a confusing site into a reference. Citability is born from pages with direct answers, lists, tables, and consistent data. llms.txt organizes the entrance; content decides the recommendation.

Bincoo takes care of both ends: it diagnoses your site's technical structure and weekly writes the content that AIs cite.

On the Essential plan, at R$ 39.00/month, you get the full technical diagnosis, optimized articles, and weekly monitoring of AI answers about your clinic.

Conclusion

• llms.txt is a Markdown map of your site's best content, designed for language models;
• It helps with discovery and reading, but it doesn't buy citations;
• List services, team, and articles with specific descriptions, and keep the file alive.

Want your clinic to be read, understood, and recommended by AIs? visit Bincoo and start with the full diagnosis.