llms.txt Template: Copyable File, Example and Guide
An llms.txt file is a markdown file at /llms.txt that gives AI systems a short, curated map of your site. The proposal at llmstxt.org requires only an H1 with your name; after it come an optional blockquote summary, optional notes, and H2 sections that list links as "- [name](url): notes". An H2 called "Optional" holds links a reader can skip when context is short. Google Search does not use llms.txt: Google said on 15 June 2026 that these files are not needed for Google Search and do not help or hurt visibility or rankings there, though keeping one for other services is fine. Copy the template below, replace every bracket, and serve it as plain text.
The template
Replace every [bracketed] value. Keep the order: H1, blockquote, notes, then the H2 link sections, with Optional last. Rename or remove the H2 sections to fit your site; only the H1 is required.
# [Company or product name]
> [One or two sentences: what you sell, who it is for, and the fact an AI assistant most needs to get right about you.]
[Optional notes in plain markdown, with no headings: what the product does and does not do, where you operate, how pricing works, and how to read the links below.]
## Docs
- [Product overview](https://www.example.com/product): [The question this page answers, in one line]
- [Pricing](https://www.example.com/pricing): [Plans, prices and what each one includes]
- [How it works](https://www.example.com/how-it-works): [The steps, in one line]
## Guides
- [Guide title](https://www.example.com/guides/guide-slug): [The question this guide answers]
## Policies
- [Privacy policy](https://www.example.com/privacy): [What you collect and how to delete it]
- [Terms](https://www.example.com/terms): [Subscription and cancellation terms]
## Optional
- [Blog](https://www.example.com/blog): [Secondary reading, safe to skip when context is short]
- [Sitemap](https://www.example.com/sitemap.xml): [Every indexable URL]
A filled example for a SaaS website
Acme Rota is a made-up shift-scheduling product, and example.com is a domain reserved for examples. The shape is what matters: one name, one summary a model can quote, notes that say what the product is not, and links grouped by what a reader needs first.
# Acme Rota
> Acme Rota is shift-scheduling software for independent restaurants and cafés. Managers build the weekly rota, staff swap shifts in the mobile app, and approved hours export to payroll. It is not a payroll or point-of-sale system.
Acme Rota is sold per location and billed monthly, with a Starter and a Pro plan. Pro adds labour-cost forecasts and multi-location reporting. Accounts are self-serve; there is no setup fee. The pages below are the canonical source for plans, features and policies.
## Docs
- [Product overview](https://www.example.com/product): What Acme Rota does for managers and for staff, with screenshots of each.
- [Pricing](https://www.example.com/pricing): Starter and Pro, what each includes, and how billing per location works.
- [Payroll exports](https://www.example.com/docs/payroll-exports): The payroll systems Acme Rota exports to and the file formats.
- [Shift swaps](https://www.example.com/docs/shift-swaps): How staff request a swap and how a manager approves it.
## Guides
- [How to build a restaurant rota](https://www.example.com/guides/restaurant-rota): A step-by-step guide for a first weekly rota.
- [Acme Rota vs spreadsheets](https://www.example.com/guides/rota-vs-spreadsheet): When a spreadsheet is enough and when it is not.
## Policies
- [Privacy policy](https://www.example.com/privacy): What staff and manager data is stored, and how to delete it.
- [Terms](https://www.example.com/terms): Monthly billing, cancellation and data export.
## Optional
- [Changelog](https://www.example.com/changelog): Dated product changes.
- [Blog](https://www.example.com/blog): Scheduling and hospitality staffing articles.
- [Sitemap](https://www.example.com/sitemap.xml): Every indexable URL.
Field by field
| Part | Required? | What to write |
|---|---|---|
| Location | Yes | /llms.txt at the root of the site. The proposal also allows one under a path, such as /docs/llms.txt, covering the URLs below it. |
| # H1 | Yes, the only required part | The name of the site or product. One line, nothing else on it. |
| > Blockquote | Optional | A short summary with the key facts needed to understand the rest of the file: what you sell, to whom, and what you are not. |
| Notes | Optional | Any markdown except headings: paragraphs or lists explaining the product and how to read the links. |
| ## Sections | Optional, zero or more | Each H2 holds a markdown list. Every item is a link, [name](url), optionally followed by a colon and a note on what the page answers. |
| ## Optional | Optional | An H2 with exactly this name. Its links are secondary; a system short on context can skip them. |
| .md pages | Optional | The proposal also suggests a clean markdown copy of each important page at the same URL plus .md (page.html.md, or index.html.md for a URL without a file name). |
How to check it before you ship it
- Fetch https://your-domain/llms.txt. It should return 200 with a text content type, not redirect to an HTML page or a 404 template.
- Read the first line. It must start with "# " followed by your name.
- Open every link. Use absolute URLs, and remove any that redirect or return an error.
- Compare the facts with the live site: prices, plan names and anything you say you do not do. A wrong price in llms.txt is a wrong price an assistant can repeat.
- If your site has many pages, generate the link sections from the same source as your sitemap, so new pages appear without a hand edit. fenn.sh/llms.txt is built that way.
What llms.txt does not do
It is a proposal, not a standard, and whether to read it is up to each AI system. It does not let a crawler past a robots.txt rule or a firewall, it does not make content that only appears after JavaScript readable, and publishing it is not evidence of visibility. Treat it as a cheap courtesy on top of the basics: retrievable pages that answer the question in their first sentences.
How Fenn helps
Fenn's readiness check looks for llms.txt on your site, alongside AI-crawler access, structured data, answer-first content, freshness and internal links. When a check fails, you get the exact change and a copyable prompt for your developer or AI coding tool. Fenn does not change your site itself; you apply the fix and mark it done.
FAQ
Where does the llms.txt file go?
At the root of your site, so it loads at https://your-domain/llms.txt. The proposal also allows a file under a path, such as /docs/llms.txt, which covers the URLs below that path.
Which parts of llms.txt are required?
Only the H1 with the name of the site or project. The blockquote summary, the notes and the H2 link sections are optional, though without links the file has little to offer.
What is the Optional section in llms.txt for?
Links that are useful but secondary. The proposal gives the H2 named Optional a special meaning: a system that needs a shorter context can skip everything in it.
Does llms.txt replace robots.txt or sitemap.xml?
No. robots.txt tells crawlers what they may fetch, and sitemap.xml lists URLs for search engines. llms.txt is a curated, readable summary with links. Keep all three.
Does llms.txt help with Google?
No. Google said on 15 June 2026 that llms.txt files are not needed for Google Search and will not positively or negatively affect visibility or rankings there. Maintaining one for other services is fine; for Google, the usual crawling, indexing and content basics still apply.
Will publishing llms.txt get my site cited by ChatGPT?
Not by itself. Reading llms.txt is up to each AI system, and citations depend on whether assistants can retrieve your pages and find a clear answer in them. Measure citations on real prompts rather than counting the file as a result.
Sources
Checked on 14 October 2026.