Free tool 04 · No sign-up
Give AI assistants a map of your site.
llms.txt is a plain-markdown index that tells AI systems what your site is and which pages matter. Fill in the form, watch the file build itself, and download a spec-correct llms.txt.
Describe your site
llms.txt · live preview
# Your site
Serve it at /llms.txt from your site root as plain text. This site publishes its own — see it live.
What is llms.txt?
llms.txt (llmstxt.org) is a proposed convention: a markdown file at your site root that gives language models a curated map of your content — one H1 with the site name, a one-line blockquote summary, then H2 sections of annotated links. Where robots.txt tells crawlers what they may not read, llms.txt tells AI systems what they should read.
Context windows are finite, so an AI assistant answering a question about your product does far better with a 40-line curated index than with your raw HTML navigation. That's the whole idea — you choose the pages that represent you.
The format, explained
The spec is deliberately minimal. Required: an H1 with the project or site name. Strongly recommended: a blockquote summary right under it. Then any number of H2 sections — Docs, Product, Pricing, Blog — each a markdown list of links in the form [title](url): description. The description after the colon is what lets a model pick the right link without fetching everything.
This generator enforces that shape as you type, so the output validates against the convention — and it serves as a live example: this site publishes its own llms.txt at clankersupport.com/llms.txt.
Where to serve it
Serve the file at /llms.txt from your site root with a text/plain or text/markdown content type. Static hosts just need the file in the public directory; Next.js apps can add a route handler; docs platforms like Mintlify generate one automatically. Keep it under a few hundred lines — it's a curated index, not a sitemap dump.
More free tools
AI support savings calculator
See how many hours and dollars AI support would save your team each month.
02CSAT calculator
Turn survey responses into a CSAT score and see how you rank against benchmarks.
03Canned response generator
Build polished, personal support replies for refunds, bugs, outages, and more.
Frequently asked questions
What is an llms.txt file?
llms.txt is a markdown file served at your site root that gives AI systems a curated index of your most important pages: an H1 site name, a one-line blockquote summary, then H2 sections of annotated links. It's a proposed convention from llmstxt.org, designed for finite context windows.
Where do I put my llms.txt file?
Serve it at the root of your domain — https://yoursite.com/llms.txt — as plain text or markdown. On static hosts, drop the file in your public folder; in Next.js, add an app/llms.txt/route.ts handler; most docs platforms can emit one automatically.
What's the difference between llms.txt and robots.txt?
robots.txt is restrictive — it tells crawlers which paths they may not fetch. llms.txt is curatorial — it tells AI systems which pages best represent your site and what each one covers. They complement each other: one sets boundaries, the other provides a reading list.
Does llms.txt actually improve AI visibility?
It's an emerging convention, not a ranking guarantee: some AI crawlers and answer engines fetch it, others don't yet. It costs one small text file, makes your site easier for any agent to summarize correctly, and adopters treat it like early sitemap.xml — cheap insurance that compounds if the convention wins.
Should I also create an llms-full.txt?
llms-full.txt is an optional companion that inlines the full text of your key pages instead of linking to them. It suits documentation sites where agents benefit from one big file. Start with llms.txt — it's the part of the convention with the widest support.
Now automate the answers
The repetitive half of support can answer itself.
Clanker Support answers from your docs, escalates to your team when it should, and installs with one script tag.