FAQ - What are llms.txt and Markdown twins, and why do AI agents read them?
do AI agents read them?
Title suggestion: AGENT - FAQ - What are llms.txt and Markdown twins, and why do AI agents read them? Shelf: Work Tags: faq, asker, intelligent-netware, llms-txt, markdown-twins, ai-agents, public-mode, ai-citation
Public answer
Question people actually type: What are llms.txt and Markdown twins, and why do AI agents read them?
When an AI assistant tries to learn about a business, it usually has to read web pages built for people: menus, images, scripts, pop-ups, and layout. A lot gets lost along the way. llms.txt and Markdown twins give machines a cleaner way in.
llms.txt is a simple text file at the root of a website, at an address like ask.yourcompany.com/llms.txt. It's a catalog: a list of your approved deposits, grouped by topic, with a one-line summary and a link for each. Think of it as a table of contents written for AI. A public room can also publish llms-full.txt, which holds the full text of every approved deposit in one file.
A Markdown twin is a plain-text copy of a deposit, published at its own address. It contains the same approved words without the page design around them. A simple way to remember it: people use the room, and agents read llms.txt and the twins.
Why do agents read llms.txt? Because it saves them work and reduces mistakes. An agent that finds a clear catalog can see what you've actually said, in your own words, and follow links to the full text. It doesn't have to piece you together from page fragments. When someone asks an AI assistant about your field, a business with clean, readable, well-organized material is easier to understand and easier to cite.
Our own public room publishes one. Open ask.intelligentnetware.com/llms.txt in any browser and you'll see the catalog for yourself.
Two things to know:
- These files exist only in public mode. An invitation-only room publishes no llms.txt and no twins, so nothing from it is indexed.
- They carry only approved deposits. Drafts never appear, because nothing reaches the room without your approval.
llms.txt is a young convention, and no one can promise how any single AI company will use it. What we can say is that it makes your approved knowledge as easy as possible for a machine to read correctly, in your words rather than someone else's summary of you.
Related questions
- [[FAQ_2026-10-02_11-public-or-invitation-only-room|Public room or invitation-only room: which should I choose?]]
- [[FAQ_2026-10-02_13-seo-and-ai-citation|Does a public Asker help with Google and with being cited by AI?]]
- [[FAQ_2026-10-02_14-network-compounding-in-real-numbers|What does network compounding look like in real numbers?]]
- [[FAQ_2026-10-02_15-what-is-syndication|What is syndication, and do I have to syndicate?]]
- [[FAQ_2026-10-02_00-INDEX|All Asker FAQs]]
_Written by Daniel Comp with GB, Director of Operations, Intelligent Netware._