Science of Apps - Launchpad · Explainer

llms.txt for startups:
does it actually help?

llms.txt is a proposed Markdown file at your site's root that summarises the site and links to key pages for AI agents, proposed by Jeremy Howard in 2024. For Google Search, it doesn't help: Google's AI optimisation guide says Search, including its generative AI features, doesn't use such files. It can still help agents and coding tools read your docs.

By , FounderUpdated 5 min read

Share

Quick answer: Does llms.txt help SEO or AI search visibility?

  1. What it is: a Markdown file at /llms.txt summarising your site and linking key pages.
  2. Who proposed it: Jeremy Howard, published 3 September 2024 (v2 updated 2026).
  3. Format: an H1 with the site name is the only required part; a summary and link lists are optional.
  4. Google Search: doesn't use it, including AI Overviews and AI Mode.
  5. Where it can help: agents and coding tools reading your documentation.
  6. What helps more: crawlable pages with clear, specific answers.

Key takeaways

  • llms.txt is a proposal, not a search-engine standard.
  • Google lists llms.txt among things you don't need for its generative AI features.
  • It isn't a crawler control: robots.txt still decides what bots may fetch.
  • For a developer product with docs, a short llms.txt is cheap and can help agents.
  • Don't spend time on it before your pages are crawlable and clear.

Every few months a new file promises to make AI engines notice you, and llms.txt is the one founders ask about most. It's a sensible idea with a narrow job, and the evidence on what it does for search is now unusually clear: Google has said in its own documentation that Search doesn't use it. This explainer covers what the file is, what Google says, where it can still help, and a short template, from the proposal itself and Google's guide, read on 17 November 2026. Jump to what it is, what Google says or when to add one.

The llms.txt proposal page titled The /llms.txt file, v2, by Jeremy Howard
The llms.txt proposal: a summary file to help agents use a website.Image: llmstxt.org · Screenshot shown for commentary, 17 Nov 2026 · source

What llms.txt is

llms.txt
Proposed byJeremy Howard, 3 September 2024 (v2 modified 10 August 2026)
PurposeHelp agents use a website: a brief summary plus links to key content
Location/llms.txt at the root, or a subpath such as /docs/llms.txt
RequiredAn H1 with the site or project name
OptionalA blockquote summary, notes, H2 sections of link lists
Companion ideaClean Markdown versions of pages at the same URL plus .md

Table: The llms.txt proposal (llmstxt.org, checked 17 Nov 2026)

What Google says

Google Search Central page Optimizing your website for generative AI features on Google Search
Google's generative AI optimisation guide: llms.txt isn't needed.Image: Google · Screenshot shown for commentary, 17 Nov 2026 · source

Google's guide to optimising for its generative AI features (updated 10 July 2026) is direct: you don't need "new machine readable files, AI text files, markup, or Markdown to appear in Google Search (including its generative AI capabilities), as Google Search itself doesn't use them". It lists llms.txt among the "AEO/GEO hacks" you can ignore, alongside chunking content and chasing inauthentic mentions.

When llms.txt is still worth adding

SituationWorth it?
Developer docs that coding agents readYes: cheap, can help agents find the right pages
An API or SDK with many pagesYes, with Markdown versions of key pages
A marketing site wanting Google visibilityNo: Google Search doesn't use it
Before your pages are crawlable and clearNo: fix the pages first

Table: When an llms.txt is worth it (editorial guidance)

A short template

Diagram of 5 steps for a short template: 1. H1; 2. Blockquote; 3. H2 "Docs"; 4. H2 "Optional"; 5. Keep it current.
A short template
  1. H1: your product name.
  2. Blockquote: one or two sentences on what it does and for whom.
  3. H2 "Docs": links to quick start, API reference, pricing, each with a short note.
  4. H2 "Optional": links agents can skip if short on space.
  5. Keep it current — an outdated map is worse than none.

What actually helps AI visibility

Checklist: What actually helps AI visibility — Crawlable pages; Clear answers; Specific facts; Consistent profiles elsewhere; Structured data that matches the page.
What actually helps AI visibility
  • Crawlable pages — allow Googlebot, OAI-SearchBot, Claude-SearchBot, PerplexityBot
  • Clear answers — one question per section, answered first
  • Specific facts — numbers, dates, sources
  • Consistent profiles elsewhere — directories, Crunchbase, launch sites
  • Structured data that matches the page — not special AI schema

The full list is in our AI search optimization checklist.

Science of Apps Launchpad homepage with the headline Launch your product. Get it found.
The Science of Apps Launchpad: a crawlable, independent profile.Image: Science of Apps · © Science of Apps · source

Ready to be found?

How we know

The format and history come from the llms.txt proposal (llmstxt.org); Google's position from its generative AI optimisation guide (updated 10 July 2026) and AI features documentation; crawler rules from OpenAI's and Perplexity's documentation. All read on 17 November 2026.

Key terms, defined

  • llms.txt: a proposed Markdown summary file for AI agents.
  • robots.txt: a file telling crawlers what they may fetch.
  • Agent: software that uses websites on someone's behalf.
  • Chunking: splitting content into small pieces, which Google says isn't needed.
  • Markdown: a plain-text formatting syntax.

Frequently asked questions

What is llms.txt?

A proposal, published by Jeremy Howard in September 2024, to put a Markdown file at /llms.txt that gives AI agents a brief summary of a site and links to its most useful pages, ideally with clean Markdown versions of those pages.

Does Google use llms.txt?

Not for Search. Google's guide to optimising for generative AI features says you don't need machine-readable files, AI text files or Markdown to appear in Google Search, including its AI features, because Google Search doesn't use them.

Does llms.txt help with ChatGPT, Claude or Perplexity?

Their crawler documentation centres on robots.txt and their bots, not llms.txt. Being crawlable by OAI-SearchBot, Claude-SearchBot and PerplexityBot, with clear answers on your pages, is what their published guidance supports.

Is llms.txt the same as robots.txt?

No. robots.txt tells crawlers what they may fetch. llms.txt is a summary and map of your content for agents; it doesn't grant or block access.

Should my startup add an llms.txt?

If you have developer documentation that agents and coding tools read, a short, maintained llms.txt is cheap and may help them. For search visibility, spend the time on crawlable, specific pages first.

What goes in an llms.txt file?

Per the proposal: an H1 with your site or project name (the only required part), an optional one-paragraph summary in a blockquote, optional notes, and H2 sections listing links to key pages with a short description of each.

Sources

  1. Jeremy Howard — The /llms.txt file, v2 (llmstxt.org; published 3 Sep 2024, modified 10 Aug 2026)
  2. Google Search Central — Optimizing your website for generative AI features on Google Search (updated 10 Jul 2026; checked 17 Nov 2026)
  3. Google Search Central — AI features and your website (checked 17 Nov 2026)
  4. Perplexity — Perplexity crawlers (checked 17 Nov 2026)
  5. OpenAI — Overview of OpenAI crawlers (checked 17 Nov 2026)
  6. Checked November 17, 2026. Figures that change — fees, salaries, prices — are reviewed every six months; if something is out of date, tell us.
Pass it on

Know someone who'd enjoy this?

Send it to them — one tap, no sign-up. Good ideas travel best between friends.

Keep exploring

Where to go from here

Join free

Get the ideas, not the noise.

Sign up with Google and we’ll send you short UX psychology tips by email — why people tap, stall and leave, and what to do about it. Unsubscribe anytime. Prefer something to keep? Get the free Laws of UX guide as a PDF.

Get the free Laws of UX PDFFree · no spam · unsubscribe anytime
Free, forever

Launch day ends. Your listing shouldn't.

What helps more than a text file: a free, crawlable Launchpad profile describing what you do in plain words, plus a product test that checks how findable you are on Google and AI assistants.

Free PDF · 74 pages · 2 editions

The Laws of UX, in one field guide

60 research-backed principles in 6 categories — what each one means, why it works on the human brain and how to use it on your own screens. Written by Sushant Singh, founder of Science of Apps, who has scaled consumer apps to lacs of users and spent 10+ years leading product across startups and digital businesses in the US, EU & India.

Pick your edition
Free PDF · 90 pages · 2 editions

100+ Subscription Paywalls, decoded

110 real paywalls from top apps, taken apart, rated out of 5 and decoded with the Laws of UX — plus the frameworks, copy and checklists to ship your own. By Sushant Singh, founder of Science of Apps.

  • 110 paywalls rated
  • 34 Laws of UX
  • 12 frameworks

Which paywall scored 4.8 out of 5, and which scored 2.6? It’s all inside.

Pick your edition
Free PDF · 188 pages · 2 editions

Unmistakable: the logo book

How the world’s greatest logos are made, chosen and kept — and how to make yours. 98 short lessons and 19 true stories for founders, designers and the curious. By Sushant Singh, founder of Science of Apps.

  • 98 lessons
  • 19 true stories
  • 60+ real logos

Why did a $35 sketch become the Swoosh, and why did Gap lose its nerve in six days? It’s all inside.

Pick your edition