What goes in an llms.txt file?
The llmstxt.org proposal sets a simple Markdown layout, in this order:
- A heading with the site’s name
- A short summary in a blockquote
- Optional notes on how to read the site
- Sections of links, each with a short description
- An “Optional” section for pages a tool can skip when it’s short on space
How is llms.txt different from robots.txt and a sitemap?
| File | Job |
|---|---|
| robots.txt | Says which crawlers may visit which pages |
| sitemap.xml | Lists every page for search engines |
| llms.txt | Summarizes the site and points to its key pages |
The robots.txt format is a published internet standard that well-behaved crawlers follow. An llms.txt file is guidance a tool may read or ignore.
Do AI assistants read llms.txt?
An llms.txt file is read when a tool chooses to read it. Documentation sites publish one so coding assistants can find their docs, and anyone can paste one into an assistant directly. Which assistants fetch it on their own, and what they do with it, is up to the companies that run them. Google says its AI features need no such file. The file is cheap to keep, so treat it as a small addition, not a strategy.
How do I make one?
- Write the site’s name as a heading.
- Add two or three sentences on what the business does, for whom, and where.
- List the main pages, each with a one-line description.
- Save it as llms.txt at the root of the site, in plain text, and update it when pages change.
Does reddysite.com have one?
reddysite.com publishes its own llms.txt, with what reddysite is, its prices and its main pages. How reddysite builds for search and AI lists what every site carries, and whether to block AI crawlers covers the file that does control access.