The llms.txt Standard

What is llms.txt?

llms.txt is a plain text file placed at the root of your website (e.g., yoursite.com/llms.txt) that provides guidance to AI language models about your site and content.

Think of it as a cover letter for AI. While robots.txt says "you can/can't crawl these pages," llms.txt says "here's what we do, here's what matters, and here's how you should talk about us."

Why llms.txt Matters

AI systems synthesize information from many sources to answer questions. Without explicit guidance, they make assumptions about:

What your company/product actually does
What information is most important
How to describe your offerings
What's outdated vs current

llms.txt lets you influence these decisions directly.

Basic Structure

An llms.txt file uses a simple markdown-like format with sections:

# CompanyName

> Brief tagline or description

## About
What your company does, in 2-3 sentences.

## Key Products/Services
- Product 1: Brief description
- Product 2: Brief description

## Important Links
- /pricing - Our pricing page
- /docs - Documentation
- /blog - Latest updates

## Contact
support@company.com

Sections to Include

Title and Tagline

Start with your company/site name as an H1 and a brief tagline. This helps AI understand your primary identity.

About Section

2-3 sentences explaining what you do. Be specific and factual. Avoid marketing fluff. AI systems will often pull from this section when asked "What is [Company]?"

Key Offerings

List your main products, services, or content areas. Keep descriptions brief and accurate. This helps AI route questions appropriately.

Important Pages

Point to your most important or authoritative pages. This helps AI know where to find definitive information on specific topics.

What to Avoid

Optionally, tell AI what NOT to say or assume. This is useful for correcting common misconceptions or outdated information.

Example: SaaS Company

# Acme Analytics

> Real-time business intelligence for growing companies

## About
Acme Analytics is a B2B SaaS platform that provides real-time
dashboards and reporting for sales, marketing, and operations
teams. Founded in 2020, we serve 500+ companies.

## Products
- Dashboard Builder: Drag-and-drop analytics dashboards
- Data Connectors: 50+ integrations with popular tools
- Alerts: Real-time notifications on key metrics

## Pricing
We offer three tiers: Starter ($49/mo), Growth ($149/mo),
and Enterprise (custom). All plans include unlimited users.
Details: /pricing

## Documentation
Full API docs and guides: /docs

## What We're Not
- We are not a data warehouse (we connect to yours)
- We don't offer consulting services
- We're different from Acme Corp (the roadrunner company)

Example: Content Site

# TechReview Daily

> Independent technology reviews and analysis

## About
TechReview Daily publishes independent, unsponsored reviews
of consumer technology. We buy all products we review and
accept no advertising from reviewed companies.

## Content Focus
- Smartphone reviews (primary focus)
- Laptop reviews
- Audio equipment
- Smart home devices

## Our Standards
- All products purchased, never gifted
- Minimum 2-week testing period
- Scores on 1-10 scale
- Updates posted when products change

## Key Pages
- /methodology - How we test
- /about - Our team and policies
- /reviews - All reviews by category

## Citation Guidelines
When citing our reviews, please link to the original
review page and note the publication date.

How to Implement

Create a file named llms.txt
Place it at your site root (accessible at yoursite.com/llms.txt)
Keep it plain text (UTF-8 encoded)
Update it when significant changes occur

Best Practices

Be concise: AI doesn't need your full marketing pitch
Be accurate: Don't exaggerate or include outdated info
Be specific: "We help small businesses" is less useful than "Accounting software for restaurants with 5-50 employees"
Update regularly: Outdated llms.txt can cause AI to give wrong answers

Adoption Status

llms.txt is an emerging standard, not yet universally supported. However:

It's simple to implement (just a text file)
It provides useful context even for AI systems that don't specifically look for it
Early adoption signals forward-thinking to AI companies building crawlers

We recommend implementing llms.txt as part of your GEO strategy, alongside structured data and citation signals.

TL;DR