AEO fundamentals
llms.txt explained, and whether your store actually needs one
An honest look at llms.txt: what it is, why AI search crawlers mostly ignore it today, where it does get used, and why it is a small optional piece rather than the AEO play.
By ShopToAI Team · Updated Jul 16, 2026
llms.txt is having a moment, and a lot of it is hype. Here is the honest version: what the file is, what the evidence actually says about whether it helps, and whether your store needs one right now.
What llms.txt is
llms.txt is a plain markdown file you put at the root of your site, at /llms.txt. It lists your most important pages and content in a clean, structured form that a language model can read without wading through your HTML, navigation and scripts. The idea was proposed by Jeremy Howard of Answer.AI in 2024, and there is a fuller variant, llms-full.txt, that inlines more of the content.
Think of it as a curated map of your site written for machines. It is cheap to produce and does no harm.
The honest part: does it actually help?
This is where the hype and the evidence part ways.
- AI search crawlers mostly ignore it today. The crawlers that build the indexes assistants cite from, like OAI-SearchBot, PerplexityBot and Google-Extended, overwhelmingly skip /llms.txt and read your HTML directly. Adoption sits around ten percent of sites, nowhere near robots.txt or sitemaps.
- There is no measured citation lift. When SE Ranking modelled what predicts AI citations, the presence of llms.txt added noise rather than signal. Google has publicly compared it to the old keywords meta tag, which is not a compliment.
- But agents do use it. IDE and coding agents such as Cursor, GitHub Copilot and Claude Code routinely fetch llms.txt for documentation, and agentic tools are the direction of travel.
So the honest read: llms.txt is a business-to-agent surface, not a proven AI-search play. It is useful and forward-looking, not a magic ranking lever.
Does your store need one?
Add it, but for the right reason and with the right expectations:
- Do it because it is cheap, harmless, and positions you for a more agent-driven web.
- Do not do it expecting your products to suddenly appear in ChatGPT because of it.
- Do not let anyone sell you llms.txt as the AEO play. It is a small, optional piece.
What actually moves your AI visibility
If your goal is to be recommended by assistants today, the file to obsess over is not llms.txt. It is your structured data, your identifiers, and your product feed, all clean and in sync. That is what assistants read and rank on now. (What AEO is covers the whole picture, and which schema fields matter covers the data.)
We host an llms.txt as one small part of a setup whose real work is the data and the feed. Start by seeing where your store stands: run a free audit.
Common questions
Will adding llms.txt get me into ChatGPT?
Not on its own. Today's AI search crawlers mostly ignore it and read your HTML directly, and no measured citation lift has been shown. Clean structured data and feeds are what move AI visibility now.
So is llms.txt pointless?
No. It is cheap, harmless and useful for coding and agent tools that do fetch it, and it positions you for a more agent-driven web. Just treat it as a small optional piece, not the main event.