Create an llms.txt file for your website in minutes. Enter your domain, edit the pages we find, and download a file that is ready to upload. Free, with no AI model in the loop: the text comes from your own pages.
llms.txt is a plain text file that sits at the root of your site. It lists your most useful pages, with a one-line description of each, so AI tools can find them without reading everything. It is a small housekeeping step, and no AI platform has confirmed that it changes what gets cited. What earns citations and a steady flow of inbound leads is the content on those pages, built around the questions your buyers ask AI search.
Untick pages you do not want. Change any title or description. Use the arrows to reorder, and change the group name to move a page under a different heading.
This updates as you edit.
The file points AI tools to your pages. What earns citations, and a steady flow of inbound leads, is the content on those pages. Omnibound builds that content from your buyer and market context, so it answers the questions your buyers ask AI search. See how it works for your company.
Book a Demollms.txt is a plain text file, written in Markdown, that you place at the root of your website, at yourdomain.com/llms.txt. It gives AI tools a short, curated map of your site: who you are, what you do, and which pages are worth reading first.
Jeremy Howard, co-founder of Answer.AI, proposed the idea in September 2024, and the format is documented at llmstxt.org. The problem it targets is simple. A language model reading a website has to dig through navigation, scripts and banners to find the useful parts, and it can only hold so much text at once. A short file of your best pages, each with a one-line description, skips that work.
Headings and bullet links. Anyone who can write a document can read it and edit it.
Link the pages that explain what you do. Listing every URL is what your sitemap is for.
No standards body has adopted it, and AI platforms are free to ignore it. More on that below.
The format is short. The generator above builds every part of it, and you can edit each line before you download.
| Part | Required? | What it does |
|---|---|---|
| H1 with your site or company name | Yes | Names the site. It is the only required element in the proposal. |
| Blockquote summary | Recommended | One or two sentences on what you do and who it is for. AI tools read this first. |
| H2 sections | Recommended | Group links by topic, such as Product, Pricing, Resources and Company. |
| Link lines | Recommended | One page per line: a title, the URL and a short description. |
| An "Optional" H2 section | No | Pages a tool can skip when it needs a shorter file. |
Here is a small example for a made-up company:
# Example Co > Example Co makes project planning software for small design studios. ## Product - [Features](https://example.com/features): What the planner does and who uses it - [Pricing](https://example.com/pricing): Plans, limits and what each plan includes ## Resources - [Planning guide](https://example.com/guides/planning): How a studio plans a project from brief to delivery ## Optional - [Careers](https://example.com/careers): Open roles at Example Co
Three files, three different jobs. One does not replace another.
| File | Its job | Who reads it | Format |
|---|---|---|---|
| robots.txt | Controls which crawlers can access which paths on your site | Search and AI crawlers | Plain text rules |
| sitemap.xml | Lists every URL you want search engines to index | Search engines | XML |
| llms.txt | Points AI tools to your best pages, with a short description of each | AI tools that choose to read it | Markdown |
Check robots.txt first. If it blocks an AI crawler, a perfect llms.txt will not get that crawler onto your pages. The free AI Visibility Checker tests crawler access, schema, rendering, robots.txt and your sitemap in one run.
The honest answer: not on its own, and nobody has shown that it does. In June 2025, John Mueller of Google said that no AI system currently uses llms.txt. Webflow's help center is just as plain: the file does not guarantee that LLMs will read your site.
So why make one? It takes about ten minutes and costs nothing. Some developer tools can read it when you point them at a site. And building it forces a useful decision: which pages on your site best explain what you do?
AI search answers a buyer's question by pulling from pages that answer that question clearly. The pages that get cited are written around what buyers actually ask: their words, their stage of research, their market. That is buyer and market context, and it is what turns a page into a citation and a citation into a steady flow of inbound pipeline. llms.txt can point to those pages. It cannot make a weak page worth citing.
The file has to load at yourdomain.com/llms.txt, in the same place as robots.txt. How you get it there depends on your platform. Pick yours:
Upload llms.txt to the public web root, the folder that holds your homepage and robots.txt. On cPanel this is usually public_html.
Several SEO plugins, including Yoast, Rank Math and All in One SEO, now offer an llms.txt setting. Check your plugin first, and edit the file there if it has one, so you do not end up with two copies.
To upload your own, use your host's File Manager or an FTP client and place llms.txt in the site root, the folder that holds wp-config.php. Do not put it in wp-content.
Shopify does not let you place files in the site root, so you use a redirect.
Wix creates an llms.txt for you once your site has a custom domain and search indexing is on. To use your version, open SEO & GEO in your dashboard, go to Tools and settings, open the llms.txt page, choose Edit file, paste your text and save.
Once you edit the file, Wix stops updating it automatically until you reset it to the default.
Recent 7.1 sites may have an llms.txt option in SEO settings. Check there first. If you do not see it, use a redirect:
HubSpot's File Manager hosts files under /hubfs/, not at your domain root, and HubSpot has said in the HubSpot Community that it is holding off on native llms.txt support. The workaround HubSpot staff suggested is a redirect.
Put llms.txt in your public folder, so the path is public/llms.txt. It is served from the root when you deploy. On Netlify, Vercel, Cloudflare Pages and GitHub Pages, commit it to the folder your build publishes as static files.
List the pages that explain what you do, who it is for and what it costs. Leave out login, checkout, legal and tag pages. The generator skips these for you.
One sentence that tells a stranger what they will learn on the page. Write it the way a buyer would phrase the question.
Product, Pricing, Resources and Company work for most sites. Rename the groups to match how your own buyers think.
Two sentences at the top: what you do and who it is for. Skip slogans and adjectives.
Update the file when you launch, rename or remove a key page. A file that lists dead pages is worse than no file.
A tidy llms.txt that points to thin pages earns nothing. Pages that answer buyer questions with specifics are the ones that get cited.
An llms.txt file tells AI tools where your pages are. Omnibound works on the pages themselves. It maps the questions your buyers ask in AI search, adds your buyer and market context, and helps your team create content that gets cited and brings in consistent inbound leads.
Book a DemoAn llms.txt file is a plain text file, written in Markdown, that sits at the root of your website. It lists your most useful pages with a one-line description of each, so AI tools can find them without reading your whole site. It was proposed in September 2024 and is documented at llmstxt.org.
Enter your domain and the generator reads your robots.txt, sitemap and homepage navigation, picks the pages most worth listing, and writes a title and description for each from the page's own tags. You edit everything, then download the file. It is free, and it does not use an AI model to write the text. You enter your email to download the file.
No. It is a proposal, not a ratified web standard. It has not been adopted by a standards body such as the W3C or the IETF, and AI platforms are free to ignore it.
Google has not said that Search uses it. In June 2025, John Mueller of Google said that no AI system currently uses llms.txt. Treat the file as low-cost housekeeping, not a ranking lever.
There is no guarantee, and no AI platform has confirmed that the file changes what gets cited. What earns citations is the content on your pages: clear answers to the questions your buyers ask, written with specifics. llms.txt only points to those pages.
Yes. They do different jobs. robots.txt controls which crawlers can access which paths, sitemap.xml lists every URL you want indexed, and llms.txt is a short, curated list of your best pages for AI tools. If robots.txt blocks an AI crawler, llms.txt will not get that crawler onto your pages.
At the root of your domain, in the same place as robots.txt, so it loads at yourdomain.com/llms.txt. A subdomain has its own root, so a file on www.yourdomain.com does not cover blog.yourdomain.com. The install section above has steps for common platforms.
Short. List the pages that explain what you do, who it is for and what it costs, usually a few dozen links at most. A full dump of every URL is what your sitemap is for. Webflow, for example, accepts uploads up to 100 KB.
llms-full.txt is a companion file that some documentation platforms publish. It puts the full text of key pages into one Markdown file so a tool can read everything in a single request. It is a convention, not part of the core llms.txt proposal. This generator creates llms.txt only.
Update it whenever you launch, rename or remove a key page, and review it at least once a quarter. A file that lists pages that no longer exist is worse than no file.
The generator picks about 30 key pages, with a handful of blog posts at most, and you can add other pages one at a time. If your site has no sitemap, it falls back to robots.txt and the links in your homepage navigation.
When you download your file, we save your email and website address so we can send occasional emails about AI search, as the line under the download button says. You can unsubscribe at any time. The pages we read to build your file are the public pages on your site.