
llms.txt: What It Is, How to Generate One, and Why AI Crawlers Look for It
llms.txt is a Markdown file that gives AI models a map of your site. The format, a working example, and what it actually affects.
There's a small file that more and more AI-facing sites are adding at their root: llms.txt. Here's what it is, what it does, and an honest take on whether you need one.
What llms.txt is
llms.txt is a proposed standard (from Jeremy Howard, September 2024): a plain Markdown file at yoursite.com/llms.txt that gives language models a curated, human-readable map of your site — what you do, what's important, and where to find it. Think of it as robots.txt's cousin: robots.txt tells crawlers what they may fetch; llms.txt tells models what's worth reading.
Why LLMs like Markdown
Markdown is close to plain text, dense with meaning per token, and structurally simple — headings, lists, links. LLMs parse it more reliably than HTML full of navigation and scripts. That's the whole idea: your value proposition and key resources in a format models ingest cleanly.
The format
- H1: your site or product name.
- A blockquote: one or two sentences describing what you offer.
- H2 sections (e.g., Docs, Product, Pricing) with Markdown links — each with a short description of what the target contains.
Keep it under a page. Curated, not exhaustive: an llms.txt with 500 links is a sitemap wearing a costume, and models will skim it the way humans do.
A working example
# AEO Pulse Labs > AI search visibility platform: monitor brand mentions and citations across ChatGPT, Perplexity, Gemini and Claude; track AI crawlers; create AI-ready content. ## Product - [AI Visibility Tracking](https://aeopulselabs.io/visibility): See how often AI engines cite your brand. - [Free AI Readiness Grader](https://aeopulselabs.io/grader): Score any URL for AI search in 60 seconds. ## Guides - [What is AEO?](https://aeopulselabs.io/blog/what-is-aeo): Answer Engine Optimization explained.
Does it affect AI citations? The honest answer
No engine has confirmed llms.txt as a ranking or citation factor — and Google has publicly downplayed it. Treat it as context, not a lever: it's cheap, takes an hour, and helps models that choose to read it understand your site faster. It is not a substitute for the things that actually drive citations: crawl access for AI bots, clean content, schema markup, and answer-first structure.
Common mistakes
- Dumping every URL instead of the 10-20 that matter.
- No descriptions on links.
- Letting it go stale when you ship new pages.
- Blocking AI crawlers in robots.txt while advertising an llms.txt — the map is useless if the territory is fenced.
- Forgetting to link it from your site.
How to validate
Fetch your file like a model does: curl https://yoursite.com/llms.txt and confirm it returns 200 with correct Markdown. AEO Pulse Labs' Site Audit checks llms.txt automatically — presence, freshness, and whether your robots.txt lets AI crawlers actually reach the pages it points to. You can also see how the Site Audit fits into the full AI-readiness workflow.
Written by the AEO Pulse Labs team. Our Site Audit validates your llms.txt, your robots.txt AI-crawler access, and the schema AI engines need to cite you.
