Most contractors have heard of robots.txt. That file tells Google what to crawl and what to skip. llms.txt does something different. It tells AI models what your website is about, what pages matter, and how to understand your business. The standard was proposed by Jeremy Howard at answer.ai in July 2024, and it is starting to show up on sites from Anthropic, Cloudflare, FastAPI, and a growing list of companies that want AI engines to read their content accurately.
For a trades business, the case is straightforward. An llms.txt file costs nothing, takes 10 minutes to set up, and aligns with where AI search is heading. No AI model has confirmed that it reads llms.txt during a live search session. But the file is structured exactly the way AI models prefer to receive information, and the companies building those models are the ones adopting it.
Here is what llms.txt is, how it works, and how a trades business can set one up without hiring a developer.
What llms.txt Actually Is
llms.txt is a plain text file written in markdown. It lives at the root of a domain, at the address yoursite.com/llms.txt. The format follows the same structure as a project README file.
A basic llms.txt file has four parts:
- A title starting with
# - A short description starting with
> - Links to key sections, each starting with
## - Optional links to detailed subpages
The spec lives at llmstxt.org, where Jeremy Howard published the original proposal. The idea is straightforward. AI models read websites by scraping HTML, which is full of navigation menus, cookie banners, footer links, and boilerplate that drowns out the actual content. llms.txt strips all of that away and hands the model a clean, prioritized summary.
How llms.txt Differs From robots.txt and Schema Markup
robots.txt controls crawling. It tells search engines which pages to index and which to ignore. It does not provide any information about what those pages contain.
Schema markup, or structured data, is the opposite. It provides detailed machine-readable information about a specific business. A LocalBusiness schema tag on a plumber's website tells Google the business name, address, phone number, services, and hours. Adding LocalBusiness schema is the single most direct way to help AI engines read business data accurately.
llms.txt sits between the two. Unlike robots.txt, it does not control crawling. Unlike schema, it does not provide structured fields. What it gives an AI model is a roadmap. Here are the important pages. Here is what they are about. Here is where to start.
Think of it as a table of contents for an AI model that has never seen your site before. If schema markup is the business card, llms.txt is the cover page of the brochure.
Should a Trades Business Create One?
The case for creating an llms.txt file is not that it will immediately boost AI search rankings. No AI model has publicly confirmed that it reads llms.txt files during live searches. The case is that it takes 10 minutes, costs nothing, and the organizations pushing the standard are the same ones building the AI models that trades businesses want to show up in.
The AEO vs SEO guide covers how AI search engines pull from different signals than traditional Google rankings. AI models look for clean, well-structured content that directly answers questions. llms.txt is designed to deliver exactly that. If the standard gains traction and AI models start reading it during inference, the businesses that already have one will be ahead. If it does not gain traction, the file costs nothing and sits harmlessly at the root of the domain.
How to Create an llms.txt File for a Trades Website
The file is plain text. No coding required. Open a text editor and follow this structure.
1. Write the title.
Start with the business name and trade.
# Smith Plumbing LLC
2. Write a one-sentence description.
> Smith Plumbing is a licensed plumbing company serving Bergen County, NJ. Residential repairs, emergency service, water heater installation, and drain cleaning.
3. Link the key pages.
List the four or five pages that matter most to a potential customer or an AI model trying to understand the business.
## [Services](/services)
Full list of plumbing services with pricing ranges.
## [Service Areas](/service-areas)
Cities and counties served in Bergen and Passaic County.
## [About](/about)
Company history, licensing, and insurance information.
## [Contact](/contact)
Phone number, email, hours, and service request form.
4. Save and upload.
Save the file as llms.txt and upload it to the root of the website. The file should be accessible at yoursite.com/llms.txt. If the site is on WordPress, this goes in the public root directory alongside robots.txt.
That is the entire process. No plugins, no paid tools, no developer time.
What llms.txt Does Not Do
llms.txt is not a ranking factor. It does not replace schema markup, Google Business Profile optimization, or building an AI visibility score. Those remain the foundation. Schema markup gives AI models structured data. Google Business Profile gives them location and review data. Service pages with clear headings give them content to cite. llms.txt adds one more layer: a plain-language summary that tells an AI model where to look and what to expect.
The file also does not guarantee that an AI model will read it. The standard is still in early adoption. As of November 2026, the companies that have implemented llms.txt are mostly tech companies and developer tools. Adoption among small businesses is low. That is actually the opportunity. If trades businesses start adding llms.txt files now, they are doing something that most competitors have not done yet. When AI models do start reading these files more systematically, the businesses that have had one in place for months will be indexed and understood faster.
FAQ
What is llms.txt?
llms.txt is a plain text file placed at the root of a website (at yoursite.com/llms.txt) that provides AI models with a summary of the site's content and links to key pages. It was proposed by Jeremy Howard at answer.ai in July 2024. The file is written in markdown and functions like a table of contents for an AI model.
Does ChatGPT read llms.txt files?
No AI model has publicly confirmed that it reads llms.txt files during live search sessions. The standard is in early adoption. However, the companies building AI models, including Anthropic and Cloudflare, are among the early adopters of the standard. Creating an llms.txt file is a low-cost, low-risk action that positions a website for when AI models do start reading it.
Is llms.txt the same as robots.txt?
No. robots.txt controls which pages a search engine is allowed to crawl. llms.txt provides a summary of what a website is about and which pages matter most. They serve different purposes and a website can have both.
How long does it take to create an llms.txt file for a trades business?
About 10 minutes. The file is plain text written in markdown. No coding or developer time is required. A contractor can write the file in a text editor and upload it to the website's root directory alongside the existing robots.txt file.
Should a trades business create an llms.txt file?
Yes. The file costs nothing to create, takes 10 minutes, and aligns with the direction AI search is heading. It does not replace schema markup or Google Business Profile optimization, which remain the foundation of AI visibility. But it adds one more layer that helps AI models understand what a business does and which pages matter.
Run your free AI visibility audit at firstcallvisibility.com to get your score across ChatGPT, Gemini, Perplexity, and Google AI Overviews. FirstCall checks all four engines and gives you a scorecard with what to fix first.