Does llms.txt Help You Show Up in ChatGPT or AI Overviews?

No, not today. An llms.txt file does not get your business into ChatGPT answers or Google AI Overviews. Google says plainly that you do not need AI text files to appear in its AI features, and in Canada Create’s own server logs the AI crawlers that do visit our pages thousands of times did not request our llms.txt file once over 40 days. The file is harmless and quick to make, so keep one if you like, but put your effort into the pages and signals these systems read.

What llms.txt is

llms.txt is a proposed standard, published on September 3, 2024 by Jeremy Howard at llmstxt.org. It suggests a plain Markdown file at the root of a website (for example example.com/llms.txt) that gives large language models a short, clean map of the site’s most useful pages. The idea is that a model reading a site at the moment someone asks a question would find a tidy summary faster than by parsing full web pages.

The proposal sets out a simple format:

• An H1 with the name of the site or project, which is the only required part
• A blockquote with a short summary
• Optional paragraphs or lists with more detail, without headings
• H2 sections, each holding a list of Markdown links with an optional note after each link
• By convention, a section called “Optional” for secondary links a model can skip

Some sites also publish an llms-full.txt with the full text of their documentation in one file. That is a common add-on, not part of the core proposal.

Does Google use llms.txt?

Google gives no sign that it does. Its guide to AI features and your website says there are no additional requirements to appear in AI Overviews or AI Mode, and that you do not need to create new machine readable files, AI text files or markup to appear in these features. The same page says the way to manage how Google crawls your site for Search is robots.txt for Googlebot.

In other words, the pages that show up in AI Overviews are pages Google can crawl, index and show a snippet from. If a page is not indexed or ranks poorly for the question, an llms.txt file will not change that.

What our own logs show

Canada Create’s own analysis: we read the access logs for canadacreate.com from September 30 to November 8, 2026 (40 days) and counted requests by user agent. The site publishes an llms.txt file and an llms-full.txt file, and its robots.txt allows the major AI crawlers on every page except a thank-you page and the terms page.

Over the same period the two files were requested 369 times in total. Almost all of those requests came from SEO audit tools, scripts and ordinary browsers, which tells you who is curious about llms.txt: marketers checking whether a site has one. The crawlers that feed AI answers went straight to the pages and to robots.txt (ClaudeBot alone read robots.txt 782 times). One site over 40 days is a small sample, and crawler behaviour can change, but it matches what the public documentation says.

Should you still add an llms.txt file?

It does no harm and it is a short job for a small business site. Some documentation platforms generate the file automatically, and a developer can point an AI coding assistant at it to read a product’s documentation. Add one if:

• You publish developer documentation, an API or a software product, where tools that read llms.txt are more common
• Your site is large and you want a short, curated list of the pages that matter most
• It costs you nothing because your website platform can generate it

Skip it, or leave it for later, if it would take time away from fixing pages that are not indexed, slow, thin or missing the answers your customers ask for. If you do publish one, keep it accurate: list your main services, locations, contact page and best guides, and update it when pages move.

What helps AI answers find and cite your business

AI Overviews and ChatGPT search pull from pages they can crawl and judge useful, so the work is mostly the same work that helps search:

• Let the crawlers in. Check robots.txt so Googlebot, Bingbot and the AI search crawlers you want can reach your public pages. Blocking a crawler is a choice you can make on purpose; doing it by accident is common.
• Answer the question in the first lines. Put a direct, self-contained answer right under each question heading, then the detail. Answer engines quote short passages that stand on their own.
• Be specific and current. Name your city, services, hours and what you do not do. Date facts that change.
• Keep one page per topic. Two pages answering the same question split the signals and confuse which one to cite.
• Keep your business details consistent. The same name, address, phone and hours on your site, your Google Business Profile and the directories AI tools read.
• Use structured data that matches the page. Organization, LocalBusiness, Article and FAQPage markup should describe exactly what a visitor can see.
• Earn mentions elsewhere. AI answers often cite directories, reviews, local news and industry lists. Being named there matters as much as your own pages.

For local businesses, our guide on how local businesses get cited in AI search goes further, and improving your website for AI Overviews covers the Google side.

What to put in robots.txt for AI crawlers

robots.txt, not llms.txt, is where you decide which crawlers may read your site. A few points to know before you edit it:

• Blocking Googlebot removes you from Google Search as a whole, including AI Overviews. Google says robots.txt for Googlebot is the control for Search.
• Google-Extended is a separate token. Per Google’s documentation, it limits use of your content for AI training and grounding in some of Google’s other AI systems; it does not control Search or AI Overviews.
• Crawlers used for AI search answers and crawlers used to train models often have different names (for example OAI-SearchBot and GPTBot). You can allow one and block the other.
• Keep private areas (admin pages, carts, thank-you pages) disallowed for everyone.

Check the current list of user agents on each company’s own documentation before you block anything, because the names change.

How to know if AI answers cite your pages

• Referral traffic. In Google Analytics 4, look at sessions from chatgpt.com, perplexity.ai, copilot.microsoft.com and gemini.google.com. AI Overview clicks arrive as ordinary Google organic traffic, so they are not separated.
• Server logs. Requests from ChatGPT-User or Perplexity-User mean a person’s question led the assistant to fetch your page at that moment.
• Ask the questions yourself. Each month, ask ChatGPT, Perplexity, Gemini and Google the ten questions your customers ask most, and note whether your site is named or linked. Use a private window and record the date, because answers vary.
• Search Console. Watch impressions for question-style queries on your guides. Rising impressions with flat clicks often means an AI Overview is answering the question on the results page.

Getting named and linked in AI answers is the focus of AI search optimization: crawl access, answer-first pages, consistent business details and mentions on the sites these engines trust.

Frequently asked questions

What is llms.txt?

llms.txt is a proposed standard, published in September 2024 at llmstxt.org, for a plain Markdown file at the root of a website that gives large language models a short summary of the site and a list of its most useful pages. It is a proposal, not a rule any search engine has adopted.

Does llms.txt help SEO?

Not for Google Search. Google says you do not need AI text files or special markup to appear in AI Overviews or AI Mode, and ranking still depends on crawlable, indexed, useful pages. An llms.txt file does no harm, but it does not replace that work.

Does Google use llms.txt?

Google gives no sign that it does. Its AI features guide says there are no additional requirements to appear in AI Overviews or AI Mode and that you do not need to create AI text files. In our own logs over 40 days, Googlebot requested pages 29,337 times and llms.txt zero times.

Does ChatGPT read llms.txt?

In Canada Create's own logs from September 30 to November 8, 2026, the crawlers behind ChatGPT search and ChatGPT's live fetches requested our pages more than 13,000 times combined and never requested our llms.txt file. They read the pages and robots.txt instead.

Is llms.txt important?

For most business websites, no. It matters more for software companies and developer documentation, where some tools read it. A small business gains more from answer-first pages, consistent business details and crawl access in robots.txt.

Can you provide an example of an llms.txt file?

A minimal file starts with an H1 such as "# Example Plumbing", a one-line blockquote summary such as "> Licensed plumbers serving Hamilton and Burlington", then H2 sections like "## Services" and "## Contact", each with a list of Markdown links to the matching pages and a short note after each link.

What should go in robots.txt for AI crawlers?

Allow the search crawlers you want to be found by, such as Googlebot, Bingbot and the AI search crawlers, and keep private areas disallowed. If you want to opt out of AI training, use the separate training tokens each company documents, for example Google-Extended or GPTBot, instead of blocking Googlebot, which would remove you from Google Search.

How do I know if AI answers cite my pages?

Check Google Analytics 4 for referrals from chatgpt.com, perplexity.ai, copilot.microsoft.com and gemini.google.com, look for ChatGPT-User or Perplexity-User requests in your server logs, and ask the main assistants your customers' top questions each month, noting whether your site is named or linked.

Where do I put an llms.txt file?

At the root of your domain, so it loads at yoursite.com/llms.txt. The proposal also allows a file inside a subfolder, such as /docs/llms.txt, to describe only that section.

Is llms.txt the same as robots.txt?

No. robots.txt tells crawlers which URLs they may visit and is followed by the major search and AI crawlers. llms.txt is an optional summary of your content with no control over crawling, and in our logs the major AI crawlers did not request it.