What is llms.txt and Why Your Business Needs One

There is a quiet shift happening on the internet right now. AI agents, the kind built into ChatGPT, Gemini, Perplexity, and dozens of vertical tools, are starting to do real work for real people. They are comparing service providers, checking availability, requesting quotes, and even booking appointments. And most businesses have zero idea it is happening.

If you run a local service business, here is the question you should be asking: when an AI agent visits your website on behalf of a potential customer, does it know what you do? Does it understand your services, your hours, your booking process? Or is it left guessing?

That is exactly the problem llms.txt was designed to solve.

What is llms.txt?

Think of llms.txt as a robots.txt file, but built specifically for large language models and AI agents. While robots.txt tells search engine crawlers which pages they can and cannot access, llms.txt tells AI systems what your business actually does, what services you provide, and how to interact with you.

It is a simple plain-text file that lives at your website's root, at yoursite.com/llms.txt. No special software required. No expensive platform. Just a text file that speaks the language AI agents already understand.

The concept comes from a growing recognition that AI agents need structured, reliable information about businesses. When an AI agent lands on a typical website, it has to scrape through marketing copy, hero images, sliders, and JavaScript-heavy pages just to figure out basic facts. That process is slow, unreliable, and often inaccurate.

With llms.txt, you hand the agent a clean, structured summary of everything it needs to know. No guessing. No scraping. Just the facts.

That number is staggering, but it also represents a massive opportunity. If you create an llms.txt file today, you are immediately in the top 4% of businesses when it comes to AI agent readiness.

What Goes in an llms.txt File?

The beauty of llms.txt is its simplicity. There are no complex schemas to learn, no validation tools required. You write it in plain text, using Markdown-style formatting. Here is what a solid llms.txt file includes:

• Business name and description:A clear, one-line summary of who you are and what you do
• Services offered:A list of your core services with brief descriptions
• Service area:Where you operate (city, state, radius)
• Business hours:When you are open and available
• Contact methods:Phone, email, contact form URL
• Booking information:How to schedule an appointment or request a quote
• Pricing guidance:Starting prices, free estimates, or pricing page links

Here is an example of what a complete llms.txt file looks like for a plumbing company:

That is it. A well-organized text file that any AI agent can parse in milliseconds. Compare that to the alternative, where an agent has to load your full website, parse JavaScript, scroll through testimonials, and try to extract the same information from marketing copy.

Why It Matters Right Now

The AI Agent Preference landscape is moving fast. Consumers are increasingly using AI assistants to find and compare service providers. When someone asks ChatGPT "find me a plumber in Dallas who can come today," the AI agent needs to quickly identify businesses that match, check availability, and present options.

Businesses without an llms.txt file are invisible in this process. Or worse, they are misrepresented because the AI had to guess based on incomplete information.

Here is what happens when an AI agent visits a business website that has no llms.txt:

1. It tries to load the homepage, which is often heavy with images and JavaScript
2. It attempts to extract the business name, services, and contact info from unstructured text
3. It may miss key services, get the hours wrong, or fail to find a booking link
4. It moves on to a competitor who made things easier

That last point is the critical one. AI agents are optimized for efficiency. They will always prefer the business that gives them clean, structured information over one that makes them work for it.

Does your site have an llms.txt file? Most do not. Run a free scan to check your score across all 4 dimensions.

How llms.txt Fits into the Bigger Picture

An llms.txt file is not a silver bullet. It is one piece of a larger AI Agent Preference framework that includes several complementary standards and protocols:

• Schema.org / JSON-LD:Structured data markup embedded in your HTML that helps AI agents understand your business type, location, reviews, and services programmatically
• MCP (Model Context Protocol):A protocol that lets AI agents interact with your business systems directly, like checking real-time availability or placing orders
• A2A (Agent-to-Agent Protocol):A framework for AI agents to communicate with each other, enabling complex multi-step transactions
• JSON-LD:The format used to embed Schema.org data in your web pages so AI agents can extract it without parsing HTML

Think of llms.txt as the front door. It is the first thing an AI agent looks for when it arrives at your domain. Once it reads your llms.txt, it knows where to find deeper integrations like your Schema.org markup, your MCP endpoints, or your agent card.

Together, these standards create a complete picture that lets AI agents confidently recommend, compare, and transact with your business.

How to Create Your llms.txt File (Step by Step)

You do not need a developer for this. If you can edit a text file, you can create an llms.txt file in under 15 minutes.

Step 1: Open a text editor

Use any plain text editor, such as Notepad on Windows, TextEdit on Mac (set to plain text mode), or VS Code if you are comfortable with it.

Step 2: Write your business summary

Start with your business name as a heading (using #) followed by a one-line description wrapped in > for a blockquote. Be specific. Do not use vague language like "we provide solutions." Say what you actually do.

Step 3: List your services

Under a ## Services heading, list every service you offer. Use bullet points. Be specific about what each service includes if possible.

Step 4: Add your service area, hours, and contact info

Create separate sections for each. Include your full hours, all contact methods, and the specific cities or areas you serve.

Step 5: Include booking and pricing details

This is where most businesses leave value on the table. If an AI agent can see that you offer free estimates, same-day service, or online booking, it is far more likely to recommend you over a competitor who does not share that information.

Step 6: Save and upload

Save the file as llms.txt and upload it to the root directory of your website, the same folder where your index.html or robots.txt lives. It should be accessible at https://yoursite.com/llms.txt.

Step 7: Test it

Open a browser and go to yoursite.com/llms.txt. If you see your content, you are good. If you see a 404 error, double-check that the file is in the right directory and that your hosting provider is not blocking .txt files.

Common Mistakes to Avoid

A few things that can reduce the effectiveness of your llms.txt file:

• Being too vague:"Full-service provider" tells an AI agent nothing. List your actual services.
• Skipping hours and availability:This is one of the most important pieces of information for booking-related queries.
• Forgetting to update it:If you add new services or change your hours, update your llms.txt. Outdated information is worse than no information.
• Making it too long:Keep it focused. This is a summary, not your entire website. Aim for 30 to 60 lines.
• Using HTML or complex formatting:Stick to plain text with Markdown-style headings and bullets. No HTML tags, no embedded images.

The Bottom Line

AI agents are not a future technology. They are here now, actively searching for businesses on behalf of consumers. The businesses that make themselves easy to find, easy to understand, and easy to book through these agents are going to win a disproportionate share of new customers.

Creating an llms.txt file takes less than 15 minutes. It costs nothing. And it puts you ahead of 96% of businesses in your industry when it comes to AI agent readiness.

The question is not whether you need one. The question is how long you can afford to go without it.

Mark Laird

Founder, GradeForAI. Creator of the AI Agent Preference Score.

Mark coined the term AI Agent Optimization and built GradeForAI to give every service business a clear measure of where they stand as AI agents reshape how consumers find and book services. More about Mark