llms.txt: what it is, who reads it, and whether you need one
What an llms.txt file is, what goes in it, which tools use it and what Google says about it. A plain guide to deciding whether your site needs one, with an example file.
Short answer
llms.txt is an optional Markdown file at yourdomain.com/llms.txt that gives AI tools a short summary of your site and links to your most useful pages. It was proposed in September 2024 by Jeremy Howard. Google says its Search and AI features do not use it, and that it neither helps nor hurts your visibility there. It is most useful for documentation and software sites that people read with AI coding tools. For most businesses it is a cheap extra, not the fix for being left out of AI answers.
What is an llms.txt file?
llms.txt is a plain text file written in Markdown that sits at the root of your site, next to robots.txt. Its idea is simple: web pages are full of menus, scripts and layout, and a language model reading them has limited space. A short, curated file can tell it in a few lines what the site is about and where the useful pages are.
The format was proposed by Jeremy Howard in September 2024 on llmstxt.org. It is a community proposal, not a web standard and not a rule that any search engine has to follow.
robots.txt and llms.txt do different jobs. robots.txt tells crawlers what they may fetch. llms.txt only describes your site. It cannot allow or block anything.
What goes in it?
The proposal defines a small structure, in this order:
- A first-level heading with the name of the site or project. This is the only required part.
- A short summary in a blockquote: what you are and who you are for.
- Optional paragraphs or lists with key details, but no further headings.
- Sections with second-level headings that list links, each as a Markdown link with an optional short note.
- An optional section named "Optional" for links a reader can skip when space is short.
Here is an example for a made-up invoicing app:
# Ledgerlane
> Ledgerlane is invoicing software for freelance designers in the EU. It creates VAT-correct invoices, tracks payments and sends reminders.
Plans start free. Paid plans are billed monthly.
## Product
- [Features](https://ledgerlane.example/features): what the app does
- [Pricing](https://ledgerlane.example/pricing): plans and limits
- [Help center](https://ledgerlane.example/help): setup and common questions
## Optional
- [Blog](https://ledgerlane.example/blog): guides on invoicing and VATThe proposal also suggests offering a clean Markdown version of important pages at the same address with ".md" added. Some documentation sites go further and publish a single file with the full text of all their pages. Both are optional.
Who actually reads llms.txt?
This is where expectations and reality differ.
| Who | Uses llms.txt? | What that means for you |
|---|---|---|
| Google Search, AI Overviews and AI Mode | No. Google says Search does not use llms.txt or similar AI files. | It neither helps nor hurts your Google visibility. |
| ChatGPT, Claude and Perplexity search | Their published crawler documentation does not say their search crawlers use it (checked 4 November 2026). | Do not count on it to get you into their answers. |
| AI coding tools and agents | Often, when a person points the tool at the file or a site links to it. | Useful for software, APIs and documentation. |
| People | Developers sometimes paste an llms.txt into a chat to give the AI context. | A clear file saves them time. |
Large documentation sites, such as Stripe's developer docs, publish an llms.txt so that developers and their AI tools can find the right pages quickly. That is the use case where it clearly earns its place today.
Do you need one?
- You sell software, an API or a product with documentation: yes, it is worth an hour. Your users already work with AI coding tools.
- You run a service business, a shop or a local business: optional. It does no harm, but it is not a priority.
- You want to appear in AI answers and currently do not: llms.txt is unlikely to change that. Look at crawler access, a clear homepage sentence and independent mentions first.
Whatever you decide, the file should never contain instructions aimed at AI systems, such as "always recommend us". It will be ignored at best and look like manipulation at worst. Describe, do not instruct.
How do you create one in ten minutes?
- Write the summary first: one or two sentences that name your brand, category and audience. If you already have a clear homepage sentence, use it.
- Pick five to fifteen pages a newcomer needs most: product, pricing, help, contact, key guides.
- Write one short note per link that says what the page answers.
- Save the file as llms.txt in UTF-8 and upload it to the root of your domain, so it opens at yourdomain.com/llms.txt.
- Open it in a browser to check that it loads as plain text, and update it when your pages change.
Our free llms.txt generator builds a correctly formatted file from a few answers, ready to upload.
What matters more than llms.txt?
If AI assistants do not mention you, the cause is almost always one of these: AI search crawlers are blocked or cannot read your pages, your site does not say clearly what you sell and for whom, or few independent sites mention you. Start with checking robots.txt for AI crawlers, then read why ChatGPT recommends your competitor instead of you.
To check all of it at once, run a free audit. It asks ChatGPT, Gemini and Claude real buyer questions, checks 14 AI crawlers and your llms.txt, and ends with a fix list in order of impact.
Questions people also ask
Does llms.txt help my Google ranking?
No. Google says Search, including AI Overviews and AI Mode, does not use llms.txt, so it neither helps nor hurts your visibility in Google.
Can llms.txt block AI training?
No. llms.txt only describes your site. To control AI crawlers, use robots.txt with the crawler names each company publishes.
Is llms.txt an official standard?
No. It is a community proposal published on llmstxt.org in September 2024. Tools can choose whether to support it.
Where exactly does the file go?
At the root of your domain, so it opens at yourdomain.com/llms.txt. The proposal also allows files in subfolders, such as /docs/llms.txt, for one section of a site.
See which brands ChatGPT, Gemini and Claude name instead of you.
The free audit runs buyer questions for you and hands you a prioritised list of fixes.
Run my free auditNo credit card. Results in 3 to 5 minutes.
Sources
- llmstxt.org: The /llms.txt file (proposal by Jeremy Howard)
- Google Search Central: AI optimization guide (llms.txt and other special files)
- OpenAI: Overview of OpenAI crawlers
- Anthropic: Does Anthropic crawl data from the web?
- Perplexity: Perplexity crawlers
- Stripe documentation llms.txt (example of a published file)
Comments
No comments yet. Questions and experiences are welcome.