What Is llms.txt? Format, Examples, Does It Help?

llms.txt is a proposed standard: a Markdown file at https://yoursite.com/llms.txt that gives AI tools — chat assistants, coding agents, answer engines — a short, curated guide to your site's most useful content.

It was proposed in 2024 by Jeremy Howard of Answer.AI. The reasoning: web pages are full of navigation, scripts and layout that waste an AI model's limited context. A clean summary with links to plain-text versions is far easier for a model to use. (What is a context window?)

What it looks like

# Acme Invoicing

> Acme sends automatic invoice reminders for freelancers. Plans start at $9/month.

Key facts: works with Stripe and PayPal; EU data hosting; free 14-day trial.

## Docs

- [Getting started](https://acme.example/docs/start.md): create an account and send a first invoice
- [Reminders](https://acme.example/docs/reminders.md): how reminder schedules work
- [API reference](https://acme.example/docs/api.md): REST endpoints and authentication

## Optional

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

The format:

An # H1 with the site or project name (required).
A > blockquote summary.
Optional free text with key details.
## sections of Markdown links, each with a short note.
An ## Optional section for things that can be skipped when context is tight.

llms-full.txt

Some sites also publish llms-full.txt: the full text of their documentation in one Markdown file, so a tool can load everything at once. Common for developer docs (Anthropic's Claude Code docs publish both). It can be large — mainly useful for docs sites.

How it differs from robots.txt and sitemap.xml

File	Audience	Purpose
`robots.txt`	Crawlers	What they may and may not crawl
`sitemap.xml`	Search engines	A complete list of URLs
`llms.txt`	AI tools	A curated, readable summary of what matters

They complement each other. llms.txt doesn't grant or deny access — that's still robots.txt. (robots.txt and sitemap.xml explained)

Does anything actually use it?

Honest answer: adoption is mixed.

Coding tools and agents often read llms.txt and .md versions of docs when given a docs URL — it's genuinely useful there.
Major search and chat products haven't committed to it as a ranking or retrieval signal.

So treat it as a cheap, low-risk improvement, not a growth hack. It takes an hour, and it's most valuable for documentation and developer products, where users point AI tools at your docs.

How to add one

Static site: put llms.txt in your public folder.

Next.js App Router: a route handler can generate it from your content:

// app/llms.txt/route.ts
export function GET() {
  const body = `# Acme\n\n> Invoice reminders for freelancers.\n\n## Docs\n- [Start](https://acme.example/docs/start.md)\n`
  return new Response(body, { headers: { 'Content-Type': 'text/plain; charset=utf-8' } })
}

Generating it from your blog or docs keeps it up to date automatically.

Also consider serving Markdown versions of key pages (e.g. /docs/start.md), which the format encourages.

Writing a good one

Lead with what you are and who it's for, in plain words.
Include the facts people ask AI about: pricing, limits, integrations, regions.
Link to the most useful pages, not every page.
Keep it current — stale pricing in an AI's answer is worse than none.

Where it fits

llms.txt is one small part of making your site easy for AI tools to understand and cite. The bigger part is clear, factual, well-structured content. (Generative engine optimization, SEO basics)

EasySpawn hosts your site on your own server, so adding routes like /llms.txt, Markdown versions of pages, or anything else is just code — with Claude Code to write it. See how it works or join the waitlist.

What Is llms.txt? A Guide for AI-Readable Websites