Free llms.txt
Generator
Enter your domain and get a ready-to-use llms.txt, the emerging standard that tells ChatGPT, Perplexity and other AI crawlers which pages matter most on your site.
One domain in. A ready to publish llms.txt out.
Enter your domain and get a starter llms.txt, built from your homepage title and your sitemap. It is the emerging file that points AI crawlers like ChatGPT and Perplexity at the pages you want them to read.
When AI describes you, it is guessing.
Ask an assistant about your site and it stitches an answer from whatever it can scrape. If your best explanation is buried under menus, marketing, and scripts, the model may summarize you wrong or cite a weak page. There is no standard spot to simply state, in your own words, what you are and which pages matter.
Your domain in. A file to paste out.
Enter your domain
Just your homepage URL. No login, no filters to learn.
We read your site
We fetch your homepage for its title and description, and your sitemap.xml for up to 25 of your real pages.
Copy your llms.txt
A ready-made markdown file with your summary and pages already listed. Tidy the labels and upload it to your root.
What these words mean
llms.txt
A plain text file at the top of your site that hands AI crawlers a short summary and a shortlist of your key pages.
served at yourdomain.com/llms.txtAI crawler
The bots behind answer engines like ChatGPT, Perplexity, and Gemini that read the web to answer people's questions.
Perplexity fetching your pagesMarkdown
A simple plain-text format that uses a hash for a heading and a dash for a list. People and models both read it easily.
a hash then Acme Analytics is a titleSitemap.xml
The machine-readable list of every URL on your site. We read it to find the pages worth putting in your file.
12 URLs pulled from your sitemapSite root
The top level of your domain, where robots.txt already lives. Your llms.txt sits right beside it, not in a subfolder.
/llms.txt, not /pages/llms.txtProposed convention
An agreed idea that some tools choose to follow, not an official rule every engine must obey. Honest framing matters here.
some crawlers read it, others do notNot a blank template. Your actual pages.
The generator reads your live homepage and sitemap, so the file arrives already filled in with your own title, your own summary, and your own pages, ready to tidy in seconds.
- Your homepage title becomes the heading
- Your meta description becomes the one-line summary
- Up to 25 real pages from your sitemap.xml, each linked
Why it works: hand a model your own words and it repeats them. Leave it guessing and it writes its own version of you.
llms.txt for acmeanalytics.com
A starting point you shape, not a URL dump.
We list what your sitemap gives us so you are not staring at a blank file. Then you keep the handful of pages you actually want quoted and drop the rest. A short, honest file beats a wall of every URL.
- We pull your pages so you edit, not start from scratch
- Trim tag pages, pagination, and thin URLs in seconds
- Rename a label so each link reads clearly to a model
Pro tip: keep the pages you would want an assistant to cite: product, pricing, docs, and key policies. Cut the rest.
Copy it, and drop it beside robots.txt.
The result gives you the whole file with one-click copy. Save it as llms.txt and upload it to your site root, the same place your robots.txt and sitemap.xml already live. That root path is where crawlers that support it look.
- Copy the whole file straight from the result
- Works on any stack, since it is just a static text file
- Re-generate any time your pages or pricing change
Rule of thumb: if robots.txt and sitemap.xml live at your root, so does llms.txt. Same neighborhood, a different job.
Small things that make it easy.
Homepage fallback
No sitemap.xml on your site? We still build a valid file from your homepage title and description.
One-click copy
Grab the entire file from the result with a single tap, ready to paste into a text editor.
Any stack
It is a static text file, so upload it however you deploy, from a CMS to a hand-built site.
Honest by design
We flag that llms.txt is an emerging convention, not a ranking hack, so you know exactly what you are getting.
Re-generate anytime
Site changed? Run it again for a fresh file in seconds, and replace the old one.
Human readable
Open it in any editor. It is plain markdown you fully control, with nothing hidden.
How to turn a generated file into a live one.
A file is only useful once it is trimmed and published. Here is the exact read on our example, acmeanalytics.com, the way you would work through it.
llms.txt examples
Three files in the exact shape this generator writes: a title, an optional summary line, then grouped links. Sections are always Product, Docs, Blog and Company, in that order, and a heading is only written when it has something under it.
# Northtrail
> Trail running shoes tested on real trails.
## Product
- [Northtrail](https://northtrail.com/): Trail running shoes tested on real trails.# Acme Analytics
> Product analytics for small teams.
## Product
- [Pricing](https://acme.com/pricing): Plans, limits and what each tier includes.
## Docs
- [Quickstart](https://acme.com/docs/start): Install the SDK and send a first event.
- [API reference](https://acme.com/docs/api): Endpoints, auth and rate limits.# Northtrail
> Trail running shoes tested on real trails.
## Product
- [Shoes](https://northtrail.com/shoes): The current range, by terrain.
## Docs
- [Fit guide](https://northtrail.com/fit): How to size for wide feet.
## Blog
- [Carbon plates](https://northtrail.com/blog/carbon): What they do on trail.
## Company
- [About](https://northtrail.com/about): Who makes them and where.For teams that want to be read right.
Get described right, early
You do not have time to fight AI engines over how your startup is summarized. Publish a clean file so the assistants that read it draw from your words and your key pages.
Point AI at the real answers
When your guides are the accurate source, an llms.txt makes them easy for assistants to find and quote. Keep the list current and answers stay grounded in your pages.
One more file at the root
You already manage robots.txt and sitemap.xml. This is the same low-cost, root-level control for the AI layer, and honest about being a still-emerging convention.
A starter file in two minutes
No bandwidth to hand-write one. Generate it from your domain, trim it to your best pages, and publish, all without a signup.
Questions, answered honestly.
Honestly, it varies. llms.txt is a proposed convention, not an official standard every engine follows. Some tools and crawlers look for it, others do not read it yet, and none are obligated to. It costs almost nothing to publish, so the practical case is low effort now with more upside as adoption grows.
No, and we would not promise a traffic boost. llms.txt is not a Google ranking factor. Its value is in AI answers: a clean summary and curated links make it easier for assistants to describe and cite you accurately, not a search ranking trick.
You still get a valid file. When we cannot find a sitemap, we build a homepage-only version from your title and description, with your homepage linked. You can add your other pages to it by hand before you publish.
Save it as llms.txt and upload it to the root of your domain so it is served at yourdomain.com/llms.txt, the same place your robots.txt lives. That root path is where crawlers that support the convention will look for it.
Yes. Enter your domain, copy the file, and upload it like any other file on your site. It is plain text, and the result shows you exactly where it goes. Any term you do not recognise is in the glossary above.
The generator builds the file, free. The dashboard then tracks whether ChatGPT, Gemini, Perplexity, and Claude actually mention and cite your pages, how you compare to competitors, and how it trends over time, alongside the rest of the platform's 60+ SEO and AI-search tools.
Enter your domain and publish a clean source.
It takes one domain. You will get a starter llms.txt built from your homepage and sitemap, ready to trim and drop at your root, so the AI engines that read it work from your words.
Sections, and what ## Optional means
This generator writes four sections, ## Product, ## Docs, ## Blog and ## Company, in that order, and only writes a heading when it has links to put under it. The spec allows others, and one of them carries a meaning worth knowing: ## Optional marks links an assistant may skip when it is short of room. Once your file is live, run it through the llms.txt validator to check its sections, links and size. llms.txt Explained covers the format in full.
See if AI actually recommends you
The checker tells you if AI can read your site. Metric Vault tracks whether ChatGPT, Gemini, Perplexity and Claude actually mention you, how you stack up against competitors, and how it trends over time, plus the rest of the platform's 60+ SEO and AI tools.