llms.txt explained
llms.txt is a proposed convention for publishing a concise guide to your website and brand information. Support varies; adding the file does not assure indexing, AI citations, or a visibility increase.
What llms.txt is
Unlike crawler access rules in robots.txt, llms.txt is descriptive content. You can place it at the website root (yoursite.com/llms.txt) as a readable guide to useful pages. It does not control how an AI system interprets your brand.
The file contains:
- Your canonical brand description
- Key products and services
- Target audience
- Differentiators
- Authoritative links (press, reviews, documentation)
- Structured FAQs
- Contact and location information
It's essentially a cheat sheet for AI models: a single, authoritative source they can reference when building an understanding of your brand.
Why it matters
AI models crawling your website have to piece together who you are from various pages: your homepage, about page, product pages, blog, etc. This is slow, error-prone, and the model might miss important context or emphasis.
llms.txt can provide a single document that says: "Here's everything you need to know about this brand, organized exactly how you need it."
Check whether the systems you care about document support, and inspect actual source evidence before claiming that a file was used.
The llms.txt format
The file follows a simple, readable structure:
# Brand Name
> One-sentence description of what the brand does.
## About
Detailed description matching your canonical brand identity.
## Products/Services
- Product 1: Description
- Product 2: Description
## Target Audience
Who this brand serves and what problems it solves.
## Key Differentiators
What makes this brand different from competitors.
## Links
- Website: https://example.com
- Documentation: https://docs.example.com
- Press Kit: https://example.com/press
## FAQ
- Q: Common question 1?
A: Answer 1
- Q: Common question 2?
A: Answer 2
The format is designed to be both human-readable (it's just markdown) and machine-parseable. AI models can extract structured information quickly and reliably.
How Outrigger generates it
Entity generates your llms.txt file from your existing brand data:
- Pulls your canonical description
- Extracts product/service information from your brand brief
- Includes your target audience and differentiators
- Adds authoritative links (website, social profiles, press mentions)
- Generates relevant FAQs based on your keyword data
Review the facts and formatting in the generated draft before asking the website owner to publish it.
Deploying llms.txt
Place the file at yoursite.com/llms.txt: the same way you'd deploy robots.txt. How you do this depends on your platform:
- Static hosting / Custom site: Upload the file to your public directory
- WordPress: Upload to the root of your WordPress installation, or use a plugin
- Vercel/Netlify: Add the file to your
public/directory - Shopify: May require a custom route or redirect
After deployment, verify it's accessible by visiting the URL directly in your browser.
Keeping it updated
Unlike robots.txt (which rarely changes), llms.txt should be updated when your brand information changes:
- New product launch → add to Products/Services
- Rebranding → update the canonical description
- New FAQ questions emerging from Sonar → add to FAQ section
- New press coverage → add to Links section
Review the published file when your brand information changes, and check any newly generated draft before replacing it.
The bigger picture
llms.txt is one piece of a broader trend: the web is being restructured for AI consumption. Schema markup, Knowledge Panels, and now llms.txt all serve the same purpose: making it easy for machines to understand who you are.
Accurate machine-readable information can make documentation easier to inspect, but it does not establish a future ranking or recommendation advantage.
In Outrigger
Entity generates and manages your llms.txt file alongside your canonical description and schema markup.