llms.txt Generator
An llms.txt generator creates the llms.txt file: a short Markdown index at the root of your site that gives AI assistants a summary of what you do and a curated list of your most useful pages. Import your sitemap or fill in the form, then copy or download a spec-valid file, or use the built-in llms.txt validator to check one you already have.
1 Describe your site
Importing reads your sitemap through our server to work around browser cross-site restrictions. Nothing fetched is stored or logged, and everything else stays in your browser.
2 Your llms.txt
1 Paste or fetch your llms.txt
Enter a domain or the full file URL. Fetching routes through our server to work around browser cross-site restrictions, and the content isn't stored or logged. Pasting stays fully in your browser.
2 Validation results
What an llms.txt file is, and what it isn't
Web pages are built for people: navigation, cookie banners, scripts and layout wrap every paragraph of real content. A language model reading your site has to dig through all of it, often within a tight context window. llms.txt, proposed by Jeremy Howard at llmstxt.org in September 2024, is a simple fix: one Markdown file at /llms.txt that states what the site is and links straight to the pages worth reading, each with a one-line note on what it covers.
It is not a crawl-control file. Unlike robots.txt, it doesn't allow or block anything, and it isn't a ranking signal: Google's guidance on AI features says you don't need machine-readable files or AI text files like it to appear in Google Search, including AI Overviews. Where it helps is with AI assistants and agents that read it directly, such as a coding assistant pointed at your API docs. It takes ten minutes to make, which is why it's worth having even with modest adoption. Structured data does the equivalent job for search engines, and the JSON-LD Schema Generator covers that side.
The llms.txt format
- An H1 with the site or project name. The only required part.
- A blockquote summary (a line starting with
>) with the key facts a model needs to understand everything else. - Optional details: paragraphs or lists with extra context, but no headings.
- H2 sections of links, each item written as
- [Name](https://url): notes, with the notes optional. - A section named "Optional", which tells tools these links can be skipped when they need a shorter context.
llms.txt example
Here is a complete llms.txt file for a fictional analytics product. It's exactly what this generator produces from the form's sample data, so you can load the same example above and edit it into your own:
# Acme Analytics
> Acme Analytics is a privacy-first web analytics platform for small teams, with a lightweight JavaScript tracker, a REST API and no cookies.
All API endpoints require an API key from your dashboard. Pricing is per site, not per pageview.
## Docs
- [Quick start](https://example.com/docs/quick-start): Install the tracker and see your first pageviews in five minutes
- [API reference](https://example.com/docs/api): REST endpoints, authentication and rate limits
- [Pricing](https://example.com/pricing): Plans, limits and what counts as a site
## Optional
- [Changelog](https://example.com/changelog): Release notes for every version
A model reading it learns what Acme is in one sentence, knows the API needs a key before opening a single page, and can pick the API reference over the quick start for an integration question. Frost Rank publishes its own at frostrank.com/llms.txt, generated from the same tool list that drives this site.
How to use this tool
-
1
Start from your sitemap or a blank form
Enter your sitemap URL and click Import to pre-fill sections from your real pages, or fill in the site name and summary by hand.
-
2
Trim the links to what matters
Keep the pages a model most needs, such as docs, pricing and key guides, grouped under clear section headings. Move secondary pages into a section named "Optional".
-
3
Describe each link
Add a short description after each link so a model knows what the page covers without fetching it.
-
4
Download and upload the file
Download llms.txt and upload it to the root of your site so it loads at https://yoursite.com/llms.txt.
-
5
Validate the live file
Switch to the Validate tab and fetch your domain to confirm the file is served correctly and follows the spec.
Common use cases
- Giving AI coding assistants a clean entry point to your API or product documentation.
- Adding llms.txt to a client site as part of an AI-search (GEO) audit.
- Turning a sitemap with hundreds of URLs into a short list of the pages that actually answer questions.
- Running the Validate tab as an llms.txt checker on a file generated by a CMS plugin, to confirm it's really served as text, not as an HTML 404 page.
Tips and common mistakes
- Curate, don't dump. llms.txt is an index of your best pages, not a second sitemap. A sitemap import is a starting point; cut it down to what a model needs.
- Write the descriptions. A bare link tells a model almost nothing; "REST endpoints, authentication and rate limits" tells it whether to open the page.
- Use absolute URLs. A model often reads the file on its own, with no page URL to resolve
/docs/apiagainst. - Check what the server returns. Many sites answer unknown paths with their homepage or a styled 404. The validator flags a file served as HTML, and the HTTP Header Checker shows the full response.
Frequently asked questions
llms.txt is a plain Markdown file served at the root of a website (yoursite.com/llms.txt) that gives large language models a short summary of the site and a curated list of links to its most useful pages. It was proposed by Jeremy Howard at llmstxt.org in September 2024 as a way to hand AI tools a clean map of a site instead of making them parse navigation, ads and scripts.
At the root of your domain, so it loads at https://yoursite.com/llms.txt, served as text/plain or text/markdown. Upload the downloaded file there, then paste your domain into the Validate tab: it fetches the live file and checks the status code, the Content-Type and the format.
No. Google's own guidance on AI features says you don't need machine-readable files or AI text files like llms.txt to appear in Google Search, including AI Overviews, so it won't change your Google rankings. Its value is for AI assistants and agents that read it directly, for example when someone points one at your documentation. Treat it as a cheap, useful extra, not an SEO lever.
robots.txt controls access: it tells crawlers which URLs they may fetch. llms.txt doesn't block or allow anything; it's a hand-picked guide to your best content. If you want to keep AI crawlers out, that still has to happen in robots.txt.
A companion convention some documentation sites use: one file containing the full text of the docs, rather than just links to them. It isn't part of the core llms.txt proposal. This tool generates the standard llms.txt index, which is the file to start with.
Yes, it's free with no signup. The file is generated and validated entirely in your browser. Only the optional sitemap import and URL fetch route through Frost Rank's server, to get around browser cross-site restrictions, and what they fetch isn't stored or logged.
Built against the llmstxt.org proposal itself, not a summary of it, and checked against Frost Rank's own live llms.txt. The validator checks the things that actually break the file in practice: a missing H1 or summary, relative links, and a server quietly returning an HTML page instead of text.
Was this tool helpful?
Rate it and leave a comment — takes 10 seconds, and genuinely helps decide what to fix or build next.