What's Changing
Will AI recommend your company? It depends on what your website says.
Buyers now ask AI to compare vendors, and it answers by reading your website. Here's how to rank in AI search and be the company it quotes.
Read More↗llms.txt
llms.txt is a plain-text file at a website's root that gives AI systems a curated index of the site: what the company does and where the important pages are, in a format language models can read in one request.
In practice
It's the AI-era cousin of robots.txt. Where robots.txt tells crawlers what they may access, llms.txt tells AI assistants what matters: services, proof, key content, stated plainly. The format is an open proposal with a published specification [1], and sites that adopt it make it easier for AI tools to describe them accurately when a buyer asks.
The file itself is deliberately simple: Markdown, starting with the company name and a one-line description, then curated links to the pages that matter (services, key products, contact, important articles), each with a short note on what it covers. The point is to hand an AI system a clean map instead of making it reconstruct one from a sprawling site. You can see ours at byalkali.com/llms.txt.
It's one piece of AI search optimization, alongside readable page structure, direct answers, and citable data.
It's worth being honest about status: llms.txt is a proposed convention, not a standard every major AI tool has committed to reading, so its payoff today is modest and forward-looking rather than guaranteed. It's also cheap enough that the calculus still favors having one. To check whether a site has adopted it, type its domain followed by /llms.txt; most don't yet, which is precisely why an accurate one is a low-cost way to stand out while the convention is young.
Common questions
Does my website need an llms.txt file?
It's a small, cheap win rather than a requirement. If buyers might research you through AI tools, giving those tools an accurate index beats letting them guess from whatever they crawl.
What's the difference between llms.txt and robots.txt?
robots.txt tells crawlers which URLs they may access; llms.txt tells AI assistants what your site is about and where the important content lives. One is permission, the other is a guided index. They coexist; neither replaces the other.
Do ChatGPT and other AI tools actually read llms.txt?
Support is uneven and still evolving; it's a proposed convention, not a guarantee that every tool honors it today. That's why we treat it as a low-cost, forward-looking win rather than a must-have: the file is trivial to publish and keep correct, and it costs little to be early.
Source
- The llms.txt specification. llmstxt.org