An llms.txt file is the equivalent of robots.txt but for language models like ChatGPT, Claude and Perplexity. While robots.txt tells Google what to crawl, llms.txt tells LLMs who you are, what you write about and which pages matter.
It is a plain Markdown file that lives at the root of your domain. Think of it as a business card for machines that generate answers.
The standard
The format was proposed by llmstxt.org in late 2024. It is not official yet. No LLM has publicly confirmed using it. But Anthropic already has one on their own domain. And crawlers from OpenAI, Google and Perplexity already look for similar files when they visit sites.
As of mid-2025, fewer than 1,000 domains had published one. Early days.
File structure
The format is simple Markdown:
# Your Site Name
> Brief description of your site or project.
## Author
- Role: Your role
- Experience: Your background
- Contact: your@email.com
## Topics Covered
- Topic one
- Topic two
- Topic three
## Content Policy
- All content is original
- Content may be cited with attribution
## Key Pages
- [Homepage](https://yoursite.com): Main page
- [About](https://yoursite.com/about): About the author
- [Sitemap](https://yoursite.com/sitemap.xml): XML sitemap
Rules to follow:
- H1 is only the site name. Nothing else on that line.
-
Blockquote (
>) goes right below with a brief description. - H2 sections group information with bulleted lists.
-
Links use standard Markdown:
- [Name](URL): description
How I implemented mine
I run a trilingual blog (ES/EN/JA) about design and AI. My llms.txt has five sections: Author, Topics Covered, Content Policy, and three Key Pages sections (one per language).
Everything is written in English because AI crawlers process in English. Your content can be in any language but the llms.txt works best in English.
Apache setup
Two things in .htaccess:
# Serve llms.txt cleanly
RewriteRule ^llms\.txt$ /llms.txt [L]
# Force plain text with UTF-8
<Files "llms.txt">
ForceType text/plain; charset=utf-8
</Files>
robots.txt reference
Add this at the end of your robots.txt:
# LLMs context file
# https://llmstxt.org/
LLMs-Txt: https://yoursite.com/llms.txt
WordPress
You can create the file manually and upload via FTP. Yoast SEO has an auto-generate option but gives you less control.
How I validated it
I asked ChatGPT to read my file directly from the URL and analyze whether it followed the llmstxt.org standard.
The first version had issues: H1 combined with the description, links without proper Markdown formatting. ChatGPT caught both. I fixed the structure, validated again and it passed.
That is the most practical way to test yours right now. Ask ChatGPT or Claude to read your llms.txt URL and tell you if the structure is correct. If an LLM can interpret it, it works.
What to expect (and what not to)
This file will not make you appear in ChatGPT answers tomorrow. No LLM officially uses it to decide what to cite. Server logs from early adopters show AI crawlers rarely visit the file yet.
But the cost is minimal: one text file, thirty minutes, zero risk. If the standard gets adopted, you are ready. It is like having a sitemap before Google formally required one.
The worst case: nothing happens. The best case: when LLMs start looking for this file, yours is already there.
You can see mine live at shinobis.com/llms.txt.
I am a UX/UI designer with 10+ years in banking and fintech. I document the real process of integrating AI into design work at shinobis.com. Writing in Spanish, English and Japanese.
Top comments (0)