Generate a spec-compliant llms.txt for your site, or check if any URL already has one. Free, no signup.
No signup100% freeLive checker included
Check any URL for an existing llms.txt
Paste any domain to see if it publishes an llms.txt manifest. We check both standard paths and score the file against the llmstxt.org spec.
Start from a template
Loads a realistic starting point into the generator below. Edit anything before downloading.
Site information
Quick import from sitemap
Pages · 1
Page 1
Pro tips
· Place llms.txt at your site root (example.com/llms.txt)
· Also add it at /.well-known/llms.txt for discovery
· Keep descriptions concise but informative
· Update when you add new documentation pages
Generated llms.txttext
# Your Site Name
## Documentation
- [Home](/): Main landing page
---
## About This File
This llms.txt file helps AI assistants understand and navigate this documentation.
Learn more at https://llmstxt.org
Generated with Dokly's llms.txt Generator: https://www.dokly.co/tools/llms-txt-generator
Dokly
Want llms.txt automatically?
Dokly generates and updates your llms.txt as you publish — no manual work, always in sync.
llms.txt is a proposed standard that helps AI assistants like ChatGPT, Claude, and Copilot understand and navigate your documentation. It's similar to robots.txt — but designed for AI consumption.
Why do you need it?
Better AI responses: AI assistants give more accurate answers about your product when they understand your docs structure.
Improved discoverability: AI tools find and reference your documentation more easily.
Future-proofing: As AI search becomes dominant, llms.txt positions your docs for AI-first discovery.
https://yourdomain.com/.well-known/llms.txt — Alternative discovery path
Frequently asked questions
What is an llms.txt file?
llms.txt is a plain-text Markdown manifest placed at the root of a website (yourdomain.com/llms.txt) that tells large language models which pages contain your primary documentation. It was proposed by Jeremy Howard in September 2024 as a way to give AI assistants a curated entry point into a site rather than forcing them to crawl and re-render every page.
How is llms.txt different from robots.txt and sitemap.xml?
robots.txt controls which crawlers are allowed on your site. sitemap.xml is a machine-readable list of every URL. llms.txt is neither — it's a curated Markdown index of the pages you most want language models to read, with a short human-written description for each. Think of it as a table of contents written for AI, not a permission file or a full URL dump.
Which AI systems actually read llms.txt?
Adoption is still early. As of 2026, the file is being used by inference-time retrieval systems and by developer tools like Cursor, Continue, and various MCP servers. Neither OpenAI nor Anthropic have publicly committed to reading it during model training, but both companies' web-search products can and do fetch it at query time when it exists. Publishing one is a low-cost bet on AI-first discovery.
Where should the file live?
The canonical location is https://yourdomain.com/llms.txt. The spec also allows https://yourdomain.com/.well-known/llms.txt as a discovery fallback. If your docs live on a subdomain like docs.yourdomain.com, publish the file on both the root domain and the docs subdomain — each host is treated independently.
What's the difference between llms.txt and llms-full.txt?
llms.txt is the curated index — links plus short descriptions. llms-full.txt is a companion file that concatenates the actual Markdown content of every listed page into a single document, so an AI can ingest your entire documentation in one fetch. Use llms-full.txt when your docs are small enough (under a few hundred KB) that a full-corpus fetch is practical.
Is publishing llms.txt going to hurt my SEO?
No. Google's indexing pipeline ignores llms.txt entirely — it's not a robots directive and it doesn't affect canonicalization. The file has zero effect on rankings, snippets, or crawl budget. It only matters to systems that explicitly look for it.
Do I need to list every page?
No, and you shouldn't. llms.txt is meant to be a curated entry point, not a sitemap dump. Include the pages you'd hand to a new engineer on their first day: getting-started, core concepts, API reference, key tutorials. Skip changelog entries, marketing pages, and legal boilerplate. A tight 20-entry file usually beats a 500-entry dump.
How often should I update it?
Whenever your top-level information architecture changes — new product area, renamed section, deprecated feature. Individual page edits don't require an update. If you're using Dokly, the file is regenerated automatically on every publish and you never have to think about it.
Does Dokly generate llms.txt automatically?
Yes. Every Dokly project publishes both /llms.txt and /llms-full.txt at the doc site's root automatically, including on the free tier. The manifest updates in real time as pages are added, renamed, or removed — no configuration, no toggle, no upgrade.
Can I include pages from a subdomain in one llms.txt file?
Cross-origin entries are allowed by the spec, but be conservative. Only reference URLs that a language model can actually fetch anonymously, and prefer keeping each subdomain's file focused on that subdomain's content. If your docs and marketing site both matter, publish two files.
How do I check if my llms.txt is valid?
This page has a live checker at the top: paste any URL and we'll fetch both standard paths and score the response against the llmstxt.org spec (H1 heading, summary blockquote, section headings, link entries). Use it on your own site after publishing, and on competitors to see who's already ahead.
Do I need to add it to my robots.txt or sitemap?
No. llms.txt is discovered by convention (root path plus .well-known fallback), not by declaration. You don't need to reference it from anywhere else. That said, if you have a custom llms.txt discovery mechanism you're welcome to link it from your homepage's <head> or from your sitemap — neither is required.