llmstxt.studio/llms.txt
# Llmstxt > A structured guide to llmstxt.studio for AI systems. This file provides an overview of key pages and resources. ## Core Resources - [Homepage](https://llmstxt.studio/): Main landing page- [Features](https://llmstxt.studio/features): Product details- [About](https://llmstxt.studio/about): Company information- [Pricing](https://llmstxt.studio/pricing): Pricing information- [Support](https://llmstxt.studio/support): Documentation- [Contact](https://llmstxt.studio/contact): Contact information- [Audit](https://llmstxt.studio/audit): Overview- [Blog](https://llmstxt.studio/blog): Blog content- [Bot](https://llmstxt.studio/bot): Overview- [Bulk Check](https://llmstxt.studio/bulk-check): Overview- [Checks](https://llmstxt.studio/checks): Overview- [Does Llmstxt Work](https://llmstxt.studio/does-llmstxt-work): Overview- [Generator](https://llmstxt.studio/generator): Overview- [Learn](https://llmstxt.studio/learn): Overview- [Llms Txt Examples](https://llmstxt.studio/llms-txt-examples): Overview ## Optional - [Privacy](https://llmstxt.studio/privacy): Overview- [Read Cost](https://llmstxt.studio/read-cost): Overview- [Research](https://llmstxt.studio/research): Overview- [Shopify](https://llmstxt.studio/shopify): Overview- [Stats](https://llmstxt.studio/stats): Overview- [Terms](https://llmstxt.studio/terms): Overview- [Wordpress](https://llmstxt.studio/wordpress): Overview- [Ai Agents Internet Target Audience](https://llmstxt.studio/blog/ai-agents-internet-target-audience): Blog content- [Fp8 Cost Of Ai Reading Your Website](https://llmstxt.studio/blog/fp8-cost-of-ai-reading-your-website): Blog content- [Llmstxt Version 2](https://llmstxt.studio/blog/llmstxt-version-2): Blog content- [Tracking Llmstxt Adoption](https://llmstxt.studio/blog/tracking-llmstxt-adoption): Blog content- [Llm Meaning](https://llmstxt.studio/docs/llm-meaning): Documentation- [Llmstxt Ai Discoverability](https://llmstxt.studio/docs/llmstxt-ai-discoverability): Documentation- [Llmstxt Implementations](https://llmstxt.studio/docs/llmstxt-implementations): Documentation- [Structured Data Ai Readiness](https://llmstxt.studio/docs/structured-data-ai-readiness): Documentation --- *Generated with [llmstxt.studio](https://llmstxt.studio)* - 30 pages listed
- 700 tokens, under the 4,000 budget
- Generic site
- Spec-compliant: H1, summary, sections
What you get, and what to do with it.
Put it live
Save it as llms.txt at the root of your site, so it answers at yourdomain.com/llms.txt. Serve it as plain text.
Where to put it on your platformGet better descriptions
This file describes each page from its address. A free account reads the pages themselves and writes a one-line description for each. 3 AI generations, free.
Enhance with AICheck it once it is live
The AI Readiness Audit runs 41 checks on your site, including whether your llms.txt is found, parses, links to pages that answer, and stays under the size that hurts.
Optimize
How the generator works
Three steps, all of them visible in the file above. No account, no crawl of your whole site, no model call. The same code builds every file we ship and every file in our research.
It reads your sitemap
/sitemap.xml or /sitemap_index.xml, plus up to two child sitemaps. A sitemap is the one list of pages you have already told search engines about, so it is the right place to start.
It ranks your pages
Home, about, pricing, docs, products and contact score highest. Tags, feeds, images and paginated archives are dropped. Shopify stores get product, collection and policy sections; WordPress sites are recognised too.
It fits the file to the size that helps
Up to 30 links, kept under 4,000 tokens. Most generators list every URL they can find. Our benchmark says that is the one thing not to do.
Size decides whether it helps at all.
We sent an AI agent to 20 websites with 5 questions each, 400 attempts in all, and counted every token it read. A small llms.txt made a site about twice as cheap to read. A large one made it about nine times more expensive than having no file. Nothing crossed the line.
What the generator writes
One Markdown file at /llms.txt, in the shape the llms.txt specification describes. Proposed in September 2024 by Jeremy Howard of Answer.AI, it has four parts, and only the first is required.
- # Title
- The name of the site or project. The only required line.
- > Summary
- A blockquote with one or two sentences on what the site is.
- ## Sections
- Headings that group links. Each link is - [name](url): note. The note is optional.
- ## Optional
- A section by that name holds pages an AI system can skip when it is short on room.
What llms.txt is for How to structure one by hand Browse real examples
# Acme the site name, required
> Invoicing software for small trades businesses.
one-line summary
## Core Resources a section of links
- [Pricing](https://acme.com/pricing): Plans and what each includes.
- [Docs](https://acme.com/docs): Setup, invoices, payments.
## Optional skippable when short on room
- [Blog](https://acme.com/blog): Company news.llms.txt, robots.txt and sitemap.xml
Three root files, three jobs. A site can have all of them, and the audit reads all three.
| File | What it does | Who reads it | Format |
|---|---|---|---|
| llms.txt | A short index of the pages that matter, with a note on each. | AI systems and agents that read your site directly. | Markdown: a title, a summary, sections of links. |
| robots.txt | Tells crawlers which paths they may and may not fetch. | Search engine and AI crawlers, by user agent. | Plain text directives, RFC 9309. |
| sitemap.xml | Lists every page you want discovered, with change dates. | Search engines. This generator, to find your pages. | XML, with optional child sitemaps. |
How to publish it
The file has to answer at /llms.txt on your domain, as plain text. Where you put it depends on what runs your site.
| Platform | Where the file goes | Guide |
|---|---|---|
| Static site (Next.js, Astro, SvelteKit, Vite, Hugo) | Put llms.txt in the folder that is served as the site root: public/ for Next.js, Astro and Vite, static/ for SvelteKit and Hugo. It deploys with the site. | Deployment guide |
| WordPress | Upload llms.txt to the site root with your host's file manager or SFTP, beside wp-config.php. Yoast SEO can also write one for you. | WordPress guide |
| Shopify | Shopify serves an llms.txt for every store on its own. Use the generated file to compare what the store already says, and to decide what to change in the store itself. | Shopify guide |
| Webflow, Squarespace, Wix, Framer | Check whether the platform writes one for you first. If not, look for a way to upload a root file; where there is none, a redirect from /llms.txt to a hosted copy works. | Deployment guide |
| Nginx, Apache, Cloudflare Pages | Serve the file as text/plain from the document root. On Cloudflare Pages, put it in the build output directory. | Deployment guide |
llms-full.txt is a convention, not part of the spec: one file holding the full text of your site so an AI system can read it in a single fetch. Documentation tools such as Mintlify write one automatically. Our research says large files cost more than they save, so this generator writes the index only.
Already have an llms.txt?
Run the audit instead. It finds the file, parses it against the spec, fetches every link to see that it answers, and measures it against the size where a file starts to cost more than it saves. Six of the 41 checks are about llms.txt alone.
Questions
Is this llms.txt generator free?
Yes. No account, no card, no watermark you have to pay to remove. Each visitor can generate up to 20 files an hour, which is a limit on abuse rather than on you.
Do I need to sign up?
No. The file above is yours to copy or download. A free account adds the things a file alone cannot do: AI-written descriptions for each page (3 free generations), an AI Readiness Audit history for your site, and a check on whether AI search engines cite you.
What if my site has no sitemap?
The generator reads your sitemap to learn which pages exist, so without one it has nothing to work from. Most platforms make one automatically at /sitemap.xml. If yours lives somewhere else, paste the sitemap address into the field instead of your domain.
How many pages does the file include?
Up to 30, ranked by how useful they are to a reader: home, about, pricing, docs, products and contact score highest, while tags, feeds and paginated archives are left out. The file is then kept under 4,000 tokens. That is deliberate. Our benchmark found a small llms.txt makes a site about twice as cheap for an AI agent to read, and a large one makes it about nine times more expensive.
Is llms.txt an official standard?
It is an open specification, proposed in September 2024 by Jeremy Howard of Answer.AI and published at llmstxt.org. It is not a W3C or IETF standard, and no major AI provider has publicly confirmed that it reads llms.txt in production. Our registry records which of 1,000,642 domains publish one, and how each file changes over time.
Does Google use llms.txt?
No. Google Search has said it does not use AI text files like llms.txt, and its own guidance is that you do not need one for Search. The file is for AI systems and agents that read your site directly, not a ranking signal.
Will this make ChatGPT or AI search engines cite my site?
There is no evidence yet that an llms.txt changes whether AI search engines cite you. What we can measure is cost: an AI agent sent to answer questions about a site with a small llms.txt spent about half the tokens it spent without one. Keep the file small and accurate and it earns its place. Expect it to do more than that and you will be disappointed.
Is llms.txt the same as robots.txt?
No. robots.txt tells crawlers which paths they may fetch. sitemap.xml lists every page. llms.txt is a short, human-readable index that tells an AI system which pages matter and what each one is about. A site can have all three, and the audit checks all three.
What is llms-full.txt?
A convention, not part of the spec: one file holding the full text of your site or docs so an AI system can read everything in a single fetch. Documentation tools such as Mintlify produce one automatically. Our research says large files cost more than they save, so this generator writes the index only.
How often should I regenerate it?
Whenever the pages you list change, and whenever you add a page that matters. Run the generator again and replace the file. With a free account you can watch the file from your dashboard, and Pro plans check your sitemap for changes every day.
Can I edit the file before I download it?
Yes. Press Edit above the file, change any title or description, then copy or download the result. Every link stays in the spec format: a name in square brackets, the address in parentheses, and an optional note after a colon.
Can I use this for WordPress or Shopify?
Yes. The generator reads any sitemap. Shopify stores get product, collection and policy sections. For where to put the file on each platform, see the publishing table above, or our WordPress and Shopify guides.
Do you store my site data?
We fetch your sitemap, build the file, and hand it to you. Nothing is saved unless you create an account and add the site. We keep a count of runs per visitor for an hour so nobody can point the generator at one host thousands of times.