llms.txt generator and checker

llms.txt is a proposed markdown file at the root of a site: an H1 with the name, a short summary, then H2 sections of links with notes. Enter your pages to generate it, or paste an existing file to check its structure; no AI crawler is documented as reading it.

Updated

One or two sentences that tell a language model what the site is.

Paragraphs or lists with any markdown except headings. Notes an assistant should know before it reads the links.

Section | Title | URL | note, or Title | URL, or just a URL. A section named "Optional" is written last and marks what an agent can skip.

If this box has text, the tool checks it against the structure instead of building a new file.

llms.txt

4 links in 3 sections

551 bytes · structure check passed

  • Pass · H1 with the nameAcme Docs
  • Pass · Blockquote summarypresent
  • Pass · Sections of links4 links
  • Pass · Optional sectionpresent

llms.txt

# Acme Docs

> Acme is a payments API for small online stores. These pages cover the API, SDKs and integration guides.

The API is REST over HTTPS and returns JSON. Test keys start with test_ and never move money.

## Docs

- [Quickstart](https://docs.example.com/quickstart.md): Create a key and take a first payment
- [API reference](https://docs.example.com/api.md): Every endpoint, parameter and error

## Guides

- [Webhooks](https://docs.example.com/webhooks.md)

## Optional

- [Changelog](https://docs.example.com/changelog.md): Past releases
  • Upload it as /llms.txt at the root of your site (the proposal also allows a subpath), served as plain text or markdown.
  • llms.txt is a proposal, not a standard, and its own page names no search crawler that reads it. It helps tools that choose to fetch it; it doesn't control crawling like robots.txt.

What is an llms.txt file, and what does it contain?

llms.txt is a proposal by Jeremy Howard, published on llmstxt.org in September 2024, for a markdown file at /llms.txt that gives language models a short, curated map of a site. Where a page of HTML is full of navigation and scripts, the file lists the pages worth reading with a line on each, so an assistant answering a question about your product can fetch the right ones.

The format is a fixed order of markdown parts. An H1 with the name of the project or site is the only required part. Next comes an optional blockquote with a short summary, then zero or more sections of any markdown except headings, then zero or more H2 sections, each a list of links written as [name](url) with an optional colon and notes. A section named Optional holds secondary links an agent can skip when it needs less context.

This tool builds that structure from your pages and checks the result against it. Paste an existing file to check one that's already live. It does not check that the links load.

# Name
> One-line summary

Details (no headings)

## Section

- [Page title](https://example.com/page): what it covers

## Optional

- [Changelog](https://example.com/changelog)

To control which AI crawlers may fetch your pages, which llms.txt does not do, use the robots.txt generator.

For the title and description search engines read, use the meta tag generator.

What are the parts of an llms.txt file?

Structure from the llmstxt.org proposal (read 2026-11-11)
PartSyntaxRequiredNotes
Name# Project nameYesThe only required section
Summary> A short summaryNoA blockquote right after the H1
DetailsParagraphs, listsNoAny markdown except headings
File lists## Section, then - [name](url): notesNoZero or more, each under an H2
Optional## OptionalNoLinks an agent can skip for a shorter context
Location/llms.txt at the root, or a subpath—A subpath file covers URLs under it

How to create an llms.txt file

  1. 1
    Write the name and a summarySay in a sentence or two what the site is. This is the first thing a model reads.
  2. 2
    List the pages that answer questionsOne per line as Section | Title | URL | note. Pick pages a user would ask about, such as docs, pricing and policies, not every URL.
  3. 3
    Put secondary pages in OptionalUse a section named Optional for pages an agent can skip, such as changelogs and archives.
  4. 4
    Copy it to /llms.txtSave the output as llms.txt at your site root and open https://your-domain/llms.txt to confirm it loads as text.

Frequently asked questions

Is llms.txt a standard?
No. The page that defines it calls itself a proposal to standardise on using an /llms.txt file. It has no standards-body backing, and each AI product decides for itself whether to read it.
Which AI crawlers read llms.txt?
We can't cite one. The proposal names no crawler that reads it; it says coding agents and chat assistants are expected to view or search it when pointed at a site, and it notes that documentation platforms such as Mintlify, GitBook and Wix generate it and that Chrome's Lighthouse audits for it. We haven't verified any search engine or AI company's statement that it reads the file.
How is llms.txt different from robots.txt and sitemap.xml?
robots.txt tells crawlers what they may fetch, and sitemap.xml lists URLs for search engines. llms.txt is a curated reading list written for language models at inference time, with a summary and notes on each link. The proposal says it is aimed mainly at inference rather than training, and that robots.txt, not llms.txt, governs access.
What is the Optional section for?
By convention in the proposal, it holds secondary information that an agent can skip when it needs a shorter context. Put reference-heavy or historical pages there, so the main sections stay lean.
Does the proposal have a larger companion file?
The proposal page we read does not describe one, so we don't generate it. Some tools produce an expanded file with full page text, but that's outside the format checked here.
Does CountryDevs use one?
Yes. Our own file is at https://www.countrydevs.com/llms.txt, generated from the same registries that build the pages it links to, so it can't drift from the site.

Sources

Related tools

See all 135 free tools

Need a custom dev tool for your team?

We build internal tools, dashboards, APIs, and SaaS platforms. If you can describe it, we can ship it.

Built by CountryDevs · No signup · No ads · We count visits, never what you enter