How the generator builds your file
We start from your XML sitemap (found through robots.txt, or at /sitemap.xml), collect up to 150 URLs, and skip things agents don't need: carts, logins, tag archives, search results, feeds and media files. If you have no sitemap, we fall back to the links on your homepage. We then visit up to 40 of the most important pages — shallow URLs and pages like About, Pricing, Services and FAQ come first — and read their titles and meta descriptions.
The output follows the llms.txt structure: an H1 with your site name, a blockquote summary taken from your homepage description, a short context line, then H2 sections grouped by URL path (Main pages, Blog, Products, Services…). Each link carries a short description. Large sections are trimmed and the overflow moves into an ## Optional section, which the spec reserves for links an agent can skip when its context is tight.
Make it great before you publish
- Rewrite the summary. One or two plain sentences: what you sell, to whom, where. “Organic spice brand from Kerala shipping across India” beats a slogan.
- Edit descriptions for facts. Mention prices in ₹, service cities, delivery times and differentiators — the details assistants need to recommend you.
- Cut the noise. Remove thin, outdated or duplicate pages. A focused list of 20 excellent links beats 150 average ones.
- Upload to the root. Serve it at
https://yourdomain/llms.txtas plain text (UTF-8). On WordPress, upload via your file manager or an llms.txt plugin; on Shopify, use a redirect to a file in Files; on Next.js, put it in/public. - Keep it current. Regenerate whenever you launch a product line, a major service page or a cornerstone guide.
llms.txt is a lightweight signal, not a magic switch. It works best alongside allowed AI crawlers and server-rendered content — run the AI Visibility Checker to see the full picture.
Frequently asked questions
What is llms.txt?
llms.txt is a proposed convention: a Markdown file at the root of your site that gives language models and AI agents a short, curated index of your most useful pages, with a one-line summary of what your business does.
Do ChatGPT or Google use llms.txt?
Adoption is still early and no major engine has confirmed it as a ranking signal. It costs minutes to publish, it's read by a growing number of agents and developer tools, and it forces you to articulate which pages matter — which helps every other AEO effort.
Is llms.txt a replacement for robots.txt or sitemap.xml?
No. robots.txt controls access and sitemap.xml lists every URL for crawlers. llms.txt is a human-readable, curated summary aimed at language models. Use all three.
What goes in llms-full.txt?
llms-full.txt is an optional companion containing the full text of your key pages or docs in one file, so an agent can load everything at once. It's most useful for SaaS documentation and knowledge bases.