llms.txt Generator — Create a Valid llms.txt File

Key takeaways

llms.txt is a proposed standard (Jeremy Howard, Answer.AI, September 2024), not an official one. Only the H1 is required.
Structure: # Name, > summary, optional paragraphs (no headings), then ## sections with "- [Title](url): notes" links. A section named "Optional" holds links that can be skipped.
As of October 2026 Google has said its search systems don't use llms.txt, and no major AI vendor has confirmed it as a ranking or crawling signal. Its practical use today is for coding assistants and agents that read it when pointed at a site.
It's cheap to publish and maintain. Treat it as a curated table of contents, not an SEO lever.

What is llms.txt and what format does it use?

Web pages are built for browsers: navigation, scripts, cookie banners and layout wrap the few paragraphs that matter. Language models have limited context windows and do better with clean text. The llms.txt proposal gives them a single, predictable Markdown file listing the pages worth reading, with a sentence about each. The format is deliberately simple so both people and parsers can handle it:

Part	Markdown	Required?	Purpose
Title	# Project name	Yes	Names the site or project.
Summary	> One or two sentences	Recommended	The key facts an assistant needs to understand everything below.
Details	Paragraphs and lists	Optional	Context, caveats, how to interpret the links. No headings allowed here.
Sections	## Docs, then - Title: notes	Optional	Groups of links to the pages that explain your product.
Optional section	## Optional	Optional	Secondary links a tool may skip when it needs a shorter context.

# Acme Analytics

> Acme is a privacy-first analytics tool for SaaS teams that connects website traffic to Stripe revenue.

Acme is a hosted service. It doesn't use cookies by default.

## Docs

- [Quickstart](https://acme.example/docs/install): Add the script tag and verify it
- [API reference](https://acme.example/docs/api): REST endpoints and authentication

## Optional

- [Blog](https://acme.example/blog)

A minimal valid llms.txt — the structure the generator produces.

How to create an llms.txt file

1.Write the H1 and the summary first. The summary is the line most likely to be quoted about you, so state plainly what the product is, who it's for and one distinguishing fact ("hosted", "open source", "pricing per event").
2.Add details only if they prevent misunderstanding — for example what you don't do, which plan a feature needs, or which product name replaced an older one.
3.Create sections for the pages that genuinely explain the product: install guide, core concepts, API reference, pricing, comparison pages. Ten to forty links is typical; hundreds defeats the purpose.
4.Use absolute https:// URLs. The file is often read out of context, where relative links can't be resolved. If you can, point to Markdown versions of pages (many docs platforms serve page.md).
5.Put "nice to have" links (blog, changelog, case studies) in a final Optional section.
6.Download the file and serve it at the root: https://yourdomain.com/llms.txt, as text/plain or text/markdown. Optionally also publish llms-full.txt with the full text of those pages inline.

Already have one? Use "Import an existing llms.txt" in the generator to load it, then fix whatever the checks flag. You can see a live example in VisitTrack's own llms.txt.

Does llms.txt help SEO or AI visibility?

Be realistic about it. Google representatives have said Google Search doesn't use llms.txt, and as of October 2026 none of OpenAI, Anthropic, Perplexity or Google has documented it as an input to their crawlers or answer ranking. Check your own server logs before assuming any crawler reads it. Where it does help: developer tools and agents (IDE assistants, MCP-based agents, docs-aware chatbots) often fetch /llms.txt or llms-full.txt when a user points them at a site, and several docs platforms generate it automatically. For a developer-facing product that's a real audience.

What moves AI visibility more: letting AI search crawlers in (check your robots.txt rules for AI bots), answer-first pages with clear facts, structured data (schema markup generator), and being mentioned on sites those engines already trust. Measure it rather than guessing: VisitTrack's AI crawler tracking shows whether ChatGPT, Claude or Perplexity actually request your llms.txt and which pages they fetch.

Common llms.txt mistakes

Dumping the sitemap into it. A list of 2,000 URLs gives a model nothing to prioritize. Curate.
Using headings in the details block. Only the H1 and H2 section headings are part of the format; the generator converts stray headings to bold text.
Relative or broken links. Use absolute URLs and re-check them when pages move.
Marketing copy in the summary. "The world's best analytics platform" tells a model nothing. Facts do.
Letting it go stale. Regenerate it when you ship major features, rename plans or change pricing — an outdated llms.txt is worse than none.
Blocking it in robots.txt or behind a bot-protection challenge, so the tools that want it can't read it.

Frequently asked questions

What is llms.txt?

llms.txt is a proposed standard for a Markdown file at a site's root (/llms.txt) that gives AI models a concise overview of the site and links to its most useful pages. It was proposed by Jeremy Howard of Answer.AI in September 2024 and is documented at llmstxt.org.

Is llms.txt required or an official standard?

No. It's a community proposal, not an IETF or W3C standard, and nothing requires it. Only one element is mandatory within the format itself: an H1 with the site or project name.

Does Google use llms.txt?

No. Google representatives have said Google Search doesn't use llms.txt, and it isn't part of Google's documentation for AI Overviews or AI Mode. Those features rely on the regular Googlebot index.

Do ChatGPT, Claude or Perplexity read llms.txt?

Not as a documented crawling or ranking input as of October 2026. Assistants and agents may fetch it when a user points them at a site or asks about it, and many developer tools use it, but the vendors haven't said their crawlers prioritize it.

What is the difference between llms.txt and llms-full.txt?

llms.txt is the short index: a summary plus links. llms-full.txt is a common companion convention that inlines the full text of those pages into one file, so a tool can load everything in one request. Keep llms-full.txt to pages that are useful to have in a model's context.

Where do I put the llms.txt file?

At the root of your domain, so it loads at https://yourdomain.com/llms.txt, served as plain text or Markdown. Subdirectory versions (for example /docs/llms.txt) are allowed by the proposal for sites with separate sections.

How is llms.txt different from robots.txt and sitemap.xml?

robots.txt says which URLs crawlers may fetch, sitemap.xml lists every URL you want indexed, and llms.txt is a curated, human-readable summary of the few pages that best explain your site. They complement each other rather than replace each other.

llms.txt generator: build a valid llms.txt for your site