llms.txt Generator

Create a valid llms.txt file to tell ChatGPT, Claude, Perplexity, and other LLMs how to navigate your site. Live preview, section builder, one-click download.

Site basics

Sections

Each section groups related links. Common sections: Docs, API, Examples, Guides.

Advanced

llms.txt preview


                    

Where to put the file

  • Upload the file to your site root, so it is reachable at https://your-site.com/llms.txt.
  • For richer coverage, also publish llms-full.txt with the actual page content in Markdown.
  • Use absolute URLs for every link so LLMs following the file can resolve them without your domain.

What is llms.txt?

llms.txt is a proposed standard (llmstxt.org, 2024) that gives large language models a concise, curated map of your site — the same job robots.txt does for search engine crawlers, but written for AI systems that need to reason about your content.

Machine-readable summary

A single Markdown file at /llms.txt that summarizes what your site is about and points to the URLs that matter most.

Cuts through crawler cost

LLMs have tiny context windows compared to your site. llms.txt tells them where to focus so they get accurate answers about you without crawling everything.

Growing adoption

Anthropic, Cloudflare, Mintlify, Cursor, and other major AI-facing companies now publish or consume llms.txt files.

Why it matters for SEO and AI visibility

LLMs are the new referrers. Being cited correctly by ChatGPT, Claude, Perplexity, and Google AI Overviews is now a distinct discipline from classic SEO — llms.txt is one of the few concrete controls you have.

1 Higher-quality citations in AI answers

When LLMs read a well-structured llms.txt, they cite the right pages instead of guessing. Fewer hallucinated URLs, more real referral traffic.

2 Reduces "AI SEO" ambiguity

Unlike vague E-E-A-T tuning, llms.txt is a concrete file with a spec. You control what LLMs see — and you can iterate on it week by week.

3 First-mover advantage

Most sites still do not publish llms.txt. Adopting the standard now positions your brand as AI-native and puts you ahead of competitors.

4 Complements robots.txt, sitemap, and structured data

llms.txt does not replace anything — it adds a new signal specifically for AI. Publish it alongside your existing SEO stack.

How to publish llms.txt

From generator to live file in four steps:

STEP 1

Fill in the basics

Enter your site name, a one-sentence summary, and (optionally) a paragraph of context. Keep the summary tight — LLMs skim it first.

STEP 2

Add sections and links

Group related URLs under sections like Docs, API, Examples. Each link is a title, an absolute URL, and an optional description that says why it matters.

STEP 3

Download the file

Click Download to save llms.txt. The file is plain Markdown and human-readable — you can review it in any editor before deploying.

STEP 4

Upload to your site root

Deploy the file so it lives at https://your-site.com/llms.txt. Serve it with Content-Type: text/markdown or text/plain.

Pro tip

Treat llms.txt as a living document. Update it whenever you ship a big feature, release a new API, or publish a landmark article — the same cadence you use for your sitemap.

Frequently Asked Questions

Everything you need to know about publishing llms.txt.

Is llms.txt an official web standard?
It is a proposed standard published at llmstxt.org by Jeremy Howard in 2024. It is not part of an IETF/W3C RFC yet, but it is already adopted by major AI companies including Anthropic, Cloudflare, and Mintlify. Adoption is a stronger signal than formal ratification.
Does llms.txt replace robots.txt or sitemap.xml?
No. robots.txt controls what crawlers may access. sitemap.xml lists every URL. llms.txt is different — it is a curated, human-written summary for AI systems that need to reason about your site under tight context limits. All three files serve distinct purposes and should live together.
What is the difference between llms.txt and llms-full.txt?
llms.txt is a lightweight table of contents. llms-full.txt is the same map, but with the full page content inlined as Markdown — so LLMs can answer questions without any additional fetches. Larger sites publish both; smaller sites can start with just llms.txt.
Which LLMs actually read llms.txt today?
AI-native browsers and agents that fetch pages on demand (Anthropic tools, ChatGPT with browsing, Perplexity, Cursor, various open-source agent frameworks) already look for llms.txt. Broader baked-in training-set use is less certain, but publishing costs almost nothing and the file becomes more valuable as adoption grows.
How long should my llms.txt be?
Short enough to fit comfortably in an LLM context window with room for the actual question. As a rule of thumb, aim for under 5,000 tokens (roughly 3,500 words). Prioritize your most important URLs; move edge cases to the "Optional" section.
What Content-Type should the server use?
text/markdown is the correct MIME type. text/plain is an acceptable fallback if your server does not support text/markdown. Do not serve it as text/html — that will confuse fetchers expecting Markdown.
Can I block LLMs from parts of my site instead?
That is a different job, handled by robots.txt with LLM-specific user agents (GPTBot, Claude-Web, ClaudeBot, PerplexityBot, etc.). llms.txt is a positive signal — "here is what to focus on." Use both: robots.txt for what to avoid, llms.txt for what to highlight.
Do I need to keep llms.txt in sync with my sitemap?
Not literally. The sitemap is exhaustive; llms.txt is curated. Only include the URLs an LLM should prioritize when explaining your site — usually 20 to 100 hand-picked entries, not the entire URL space.
Does publishing llms.txt hurt my SEO?
No. Google, Bing, and other traditional search engines ignore it. Only AI systems that specifically look for /llms.txt read it. There is no crawl-budget or ranking cost.