LLMs.txt Generator
Build an llms.txt from a short form, so ChatGPT, Claude and Perplexity get a clean, curated map of your site, with a summary and your key pages, ready to copy.
Site details
Key pages
Your llms.txt
llms.txt is a proposed convention for a single markdown file at the root of your site that gives AI assistants a clean, curated summary of what you offer and where the important pages are. It is not an official standard yet, but tools like ChatGPT, Claude and Perplexity increasingly read sites to answer questions, and a tidy file is far easier for them to parse than a sprawling navigation. This free generator builds a correctly formatted llms.txt from a short form: a site name, a one-line summary, an optional description and a list of key pages. The output updates live and is ready to copy. It is built for founders, marketers and developers who want to be understood clearly by AI.
How to use it
Describe your site
Enter your site name and a one-line summary of what it does. Add a longer description if you want to give more context; it is optional and becomes a short paragraph in the file.
List your key pages
Add the pages that matter most, docs, pricing, your blog, each with a title, a URL and an optional note. Use the add and remove buttons to build the list, and keep it to the pages you actually want cited.
Copy it to your site
Copy the generated markdown, save it as a plain file named llms.txt and upload it to the root of your domain so it is reachable at /llms.txt. That is where AI tools following the convention look for it.
Why it's useful
Correct markdown, by construction
The generator writes the file in the proposed format: an H1 with your site name, a blockquote summary, an optional description paragraph and a Pages section of markdown link bullets. You get the structure right without learning the spec or hand-formatting links.
A curated map, not your whole site
An llms.txt is a short, human-chosen list of the pages you most want understood and cited, not a dump of every URL like a sitemap. Choosing a handful of clear, high-value pages gives an assistant a clean signal of what your site is about.
Nothing leaves your browser
The file is built entirely client-side. Nothing you type is uploaded or stored, so you can draft an llms.txt for an unreleased product or a client site without signing in.
What llms.txt is, and how to make the most of it
A proposed convention, not an official standard
llms.txt was proposed in 2024 as a simple way to help large language models find and understand the most useful content on a website. The idea is a single markdown file at /llms.txt that opens with the site name as an H1, a one-line summary in a blockquote, an optional paragraph of context, and then one or more sections of curated links to key pages. Markdown is used because it is easy for both people and models to read, and the curated list keeps the signal high.
It is important to be clear that this is an emerging convention, not an official standard, and not something every AI tool reads today. Adoption is growing but uneven, and no search engine or assistant is obliged to use the file. Treat llms.txt as a low-cost, forward-looking addition: it does not replace your sitemap, your robots.txt or good on-page content, and it will not on its own change how you are ranked or cited. It simply offers a clean, machine-friendly summary for the tools that do choose to read it.
How llms.txt differs from a sitemap and robots.txt
These three files do different jobs and work together rather than overlapping. A sitemap.xml lists every URL you want crawled, for completeness. A robots.txt tells crawlers what they may and may not fetch, for control. An llms.txt is editorial: a short, curated summary that points an AI assistant at the handful of pages you most want it to understand, for clarity. A sitemap is exhaustive; an llms.txt is selective.
That selectivity is the point. Rather than making a model infer what matters from a hundred links, you tell it directly: here is what we do in one line, and here are the five pages that explain it. Write the summary the way you would describe your site to a person in a sentence, give each linked page a short note on what it covers, and keep the list to the pages you would genuinely want quoted in an answer.
Keep it short and current
An llms.txt earns its keep by being concise and accurate. A handful of well-chosen pages with clear notes is more useful than a long, stale list, and because it is hand-curated it needs an occasional review so the summary and links still match your site. Update it when your core pages or positioning change.
Being understood, then being found
A clear llms.txt helps an assistant describe your site accurately, but being understood is only the first step; the harder part is being the source worth citing in the first place. That still comes from genuinely useful content, published consistently and structured well, in the languages your audience actually searches in. That is where Otorank picks up: it researches, writes, publishes and tracks content built to be found across 150+ languages, so once your llms.txt gives AI a clean summary, there is substance behind it worth surfacing, in every market you serve.
They use Otorank
Questions
What llms.txt is, whether it is an official standard, where it goes, and how it differs from a sitemap.
A proposed convention for a single markdown file at the root of your site that gives AI assistants a curated summary of what you offer and links to your key pages, so they can understand and cite your site more easily.
No. It is an emerging convention proposed in 2024, not an official standard, and no AI tool is obliged to read it. Adoption is growing but uneven, so treat it as a low-cost, forward-looking addition rather than a requirement.
At the root of your domain, so it is reachable at https://yourdomain.com/llms.txt. Save the generated text as a plain file named llms.txt and upload it there.
A sitemap lists every URL for crawling, to be exhaustive. An llms.txt is a short, hand-curated summary that points AI tools at the handful of pages you most want understood, to be clear. They complement each other.
It helps AI tools that read it describe your site accurately, but it will not on its own make you ranked or cited. Being surfaced still depends on genuinely useful, well-structured content; the file simply makes that content easier to summarise.