What llms.txt actually is
It is a plain markdown file at the root of your domain, yourdomain.com/llms.txt, that describes your business and points to your most important pages, written for machines rather than people.
Think of it as the AI-era equivalent of robots.txt, except robots.txt says what crawlers may not do, while llms.txt says what your site is. It typically opens with a summary of the organisation, then lists key pages grouped by category with a line of context each.
Who actually reads it
Here is the honest position as of 2026. The llms.txt convention was proposed in 2024 and adoption is growing among documentation sites, developer tools and increasingly among businesses. Several AI crawlers and retrieval systems fetch it. But no major provider has formally committed to using it as a ranking or citation input, and you should be sceptical of anyone claiming otherwise.
What it reliably does: gives any system that fetches it a clean, unambiguous, self-authored description of your business, free of navigation, cookie banners and marketing chrome. Whether a given model weights that heavily is not publicly documented.
Why it is still worth doing
- The cost is near zero. One file, written once, updated occasionally.
- The downside is nil. It cannot hurt rankings and it is invisible to human visitors.
- You control the framing. Without it, models infer what you do from whatever text they scraped. With it, you state it plainly.
- The exercise is valuable independently. Writing a clear summary of what your business does and which pages matter forces clarity most sites lack.
- Early positioning. If adoption formalises, having one already is free. If it does not, you lost an hour.
How to write a good one
Keep it factual and specific. Open with a blockquote summarising the business in two or three sentences, what you do, who for, where. Then contact details. Then your key pages grouped sensibly, services, locations, industries, articles, each with a short description.
Avoid marketing language entirely. "Award-winning, customer-focused solutions provider" tells a model nothing. "New Orleans web design and AI search optimization agency, operating since 2001, serving attorneys, restaurants, contractors and medical practices across the greater metro" tells it everything.
Many sites also publish llms-full.txt, the same structure with fuller descriptions per page.
Honest limitations
llms.txt will not rescue a site with blocked crawlers, thin content or no structured data. It is the last five percent, not the first. If your AI visibility is poor, the cause is almost certainly one of the seven common blockers rather than a missing markdown file.
Deploy it because it is cheap and sensible. Do not let anyone sell it to you as the reason you will start appearing in ChatGPT.
llms.txt, frequently asked questions
Is llms.txt an official standard?
No. It is a community convention proposed in 2024 with growing but voluntary adoption. It is not ratified by any standards body and no major AI provider has formally committed to honouring it. Treat claims to the contrary with suspicion.
Where does the file go and what format is it?
Plain markdown at the root of your domain, at /llms.txt, served as text. Some sites also publish /llms-full.txt with expanded descriptions. No plugin is required, it is a static file.
Will llms.txt make me appear in ChatGPT?
On its own, no. Crawler access, structured data, entity consistency and genuinely useful content do the heavy lifting. llms.txt makes it easier for a system that already reaches you to understand you accurately. It is worth doing precisely because it is cheap, not because it is decisive.