llms.txt present

What this means

https://YOUR_DOMAIN/llms.txt returns 200 as plain text. It is a short, factual description of your site written for AI crawlers: what the product is, who made it, the main pages, every price point, and how to cite you. Failing means the file is missing or the host serves its 404 page in its place.

How we test it

We request /llms.txt on your domain, following at most 3 redirects. A 200 status with a text content type passes. A 404, or a 200 whose content type is text/html because the host is serving its not-found page, fails. This check carries info severity and does not change your grade.

Why it matters

This is a low-cost hint for AI crawlers; hygiene, not strategy. Crawlers that fetch it use it to describe you, and a file at the root that states the facts plainly is cheaper than hoping a model infers them from marketing copy. The Fix Kit treats it as a required part of being found, with full prices listed because a "from" price gives an answer engine nothing to quote, and it is a precondition in the launch module.

The fix

Copy templates/llms.txt to the public root of the site so it serves at /llms.txt (Next.js: public/llms.txt), fill every bracket with facts, remove the sections that do not apply, list every price point, and make sure every link in it resolves. Then confirm with curl -sI https://YOUR_DOMAIN/llms.txt | grep -iE "^(HTTP|content-type)" that it returns 200 and text/plain. The full runbook is in the Fix Kit, Module 02.

Prompt for your AI

Paste this into the coding agent that built the app. It reproduces the finding and stops before changing anything; the fix steps are in the kit.

Do not change anything yet. This is a verification pass only.
Run: curl -sIL --max-redirs 3 https://YOUR_DOMAIN/llms.txt | grep -iE "^(HTTP|content-type)"
Tell me whether the status is 200 and whether the content type is text/plain or text/html.
If it returns 200, run curl -s https://YOUR_DOMAIN/llms.txt | head -40 and show me the output.
Check whether a file named llms.txt exists anywhere in the repo and print its path, or say that none exists.
List what it would take to serve a factual llms.txt at the root, without editing anything.
Verify: the headers show HTTP 200 and a text/plain content type; if not, report what was returned and stop.

Copy the prompt