Skip to content
Black BoxPersonalization

Understand

What is llms.txt?

A Markdown file at /llms.txt. This page reads the one this site serves, next to robots.txt and the sitemap.

The proposal

llmstxt.org is a 2024 proposal by Jeremy Howard. A site may publish Markdown at /llms.txt so a model can start with a name, a summary, and a few chosen links.

The usual shape is one H1, a blockquote summary, optional notes, then H2 sections of markdown links. An optional /llms-full.txt can expand those pages into one long document. This origin does not serve /llms-full.txt.

Which AI tools read the file is unconfirmed, and the public evidence is limited. Publishing it is cheap. It is a speculative hint, not a ranking lever. A crawler that requests the URL has fetched a file. That request does not show the file changed an answer.

Access

robots.txt

Tells a polite crawler which paths it may fetch. It does not describe the site.

User-Agent: *
Allow: /
Disallow: /account
Disallow: /admin
Disallow: /api
Disallow: /white-paper/download
Disallow: /white-paper/thanks

Sitemap: https://www.blackboxpersonalization.com/sitemap.xml

Discovery

sitemap.xml

45 public URLs. A list for crawlers, not a reading guide.

  • /live
  • /blog
  • /white-paper
  • /about
  • /contact

Guidance

llms.txt

A short brief. The links are chosen, not every URL in the sitemap.

# Black Box Personalization

> Opening the black box of personalization. A public site by Neil Black about how digital metrics are collected in the browser, on the network, and at the server, and how those signals can personalize a page.

The public origin is https://www.blackboxpersonalization.com. Pages teach with the live mechanism. This file is a curated brief. It does not replace robots.txt or the sitemap.

## Start here

- [Home](https://www.blackboxpersonalization.com/): Where a metric is read in the browser, on the network, and in the stored beacon.
- [Lab](https://www.blackboxpersonalization.com/lab): Collection, understanding, and personalization, with the rule on the page.
- [About](https://www.blackboxpersonalization.com/about): Neil Black.
- [Notes](https://www.blackboxpersonalization.com/blog): Short essays.

This site’s file

Open /llms.txt

The lines below are the body served at /llms.txt. Each note names the part of the proposal that line fills.

  1. 1 · H1

    # Black Box Personalization

    One H1. The proposal uses it as the project or site name.

  2. 2 · Blank

     

    A blank line between parts.

  3. 3 · Summary

    > Opening the black box of personalization. A public site by Neil Black about how digital metrics are collected in the browser, on the network, and at the server, and how those signals can personalize a page.

    A blockquote. A short summary a model can read before it follows a link.

  4. 4 · Blank

     

    A blank line between parts.

  5. 5 · Detail

    The public origin is https://www.blackboxpersonalization.com. Pages teach with the live mechanism. This file is a curated brief. It does not replace robots.txt or the sitemap.

    Optional prose. The proposal allows notes between the summary and the link lists.

  6. 6 · Blank

     

    A blank line between parts.

  7. 7 · Section

    ## Start here

    An H2. Links under it are one group.

  8. 8 · Blank

     

    A blank line between parts.

  9. 9 · Link

    - [Home](https://www.blackboxpersonalization.com/): Where a metric is read in the browser, on the network, and in the stored beacon.

    A markdown link, then a colon and a note about that URL.

  10. 10 · Link

    - [Lab](https://www.blackboxpersonalization.com/lab): Collection, understanding, and personalization, with the rule on the page.

    A markdown link, then a colon and a note about that URL.

  11. 11 · Link

    - [About](https://www.blackboxpersonalization.com/about): Neil Black.

    A markdown link, then a colon and a note about that URL.

  12. 12 · Link

    - [Notes](https://www.blackboxpersonalization.com/blog): Short essays.

    A markdown link, then a colon and a note about that URL.

  13. 13 · Link

    - [White paper](https://www.blackboxpersonalization.com/white-paper): How the site was built.

    A markdown link, then a colon and a note about that URL.

  14. 14 · Blank

     

    A blank line between parts.

  15. 15 · Section

    ## Measurement

    An H2. Links under it are one group.

  16. 16 · Blank

     

    A blank line between parts.

  17. 17 · Link

    - [AI referrals](https://www.blackboxpersonalization.com/lab/ai-referrals): How this site classifies AI referrers and crawler user agents.

    A markdown link, then a colon and a note about that URL.

  18. 18 · Link

    - [llms.txt](https://www.blackboxpersonalization.com/lab/llms-txt): What this file is, and how it differs from robots.txt and the sitemap.

    A markdown link, then a colon and a note about that URL.

  19. 19 · Link

    - [SEO](https://www.blackboxpersonalization.com/lab/seo): What a crawler reads in the HTML.

    A markdown link, then a colon and a note about that URL.

  20. 20 · Link

    - [Sitemap](https://www.blackboxpersonalization.com/sitemap.xml): The full public URL list.

    A markdown link, then a colon and a note about that URL.

  21. 21 · Blank

     

    A blank line between parts.

Try a draft from the sitemap

The button builds a Markdown file from the 45 URLs in this site’s sitemap. Link text is the path. The result stays in the page. It does not replace /llms.txt.

Has a crawler fetched it?

The AI referrals lab logs document requests whose User-Agent matches the crawler list. Rows here are that log, limited to path /llms.txt.

No logged AI user agent has fetched /llms.txt.

AI referrals and the full crawler log

Why these picks

Reading path transitions…