Free llms.txt Generator
Create a ready-to-review llms.txt for an entire website or a specific section. No account required.
The generator reads the public sitemap and page metadata, then uses AI to select and organize the most useful pages. Every URL comes from the crawl. Review the draft before publishing it.
What is llms.txt?
llms.txt is a plain Markdown file that gives AI systems a curated guide to a website. It contains an H1 site name, a short blockquote summary, and sections linking to the most useful pages. Answer.AI proposed the format in September 2024, and the specification is published at llmstxt.org. It is intended to be selective rather than duplicate a full sitemap. This generator follows v2, released on 10 August 2026.
Some websites publish llms.txt, but no major AI vendor currently documents support for third-party files. Google Search states that its AI features do not use llms.txt. Treat it as a low-cost navigation aid, not a ranking signal. See the llms.txt entry for the evidence and the deployment playbook for serving and verification steps.
How it works
- Read the homepage to identify the site name, description, and language.
- Find pages through robots.txt and the sitemap. If no sitemap is available, use links from the homepage.
- Fetch selected pages to confirm their titles and descriptions, and remove links that no longer resolve.
- Use AI to select useful pages, group them into sections, and write the summary and descriptions in the site’s language. If AI drafting is unavailable, assemble the file with predefined rules.
What the output looks like
This example uses a fictional analytics product. Your result follows the same structure and uses your website’s language:
# Northwind Analytics > Northwind Analytics is a self-hosted product analytics platform with HTTP event ingestion, ClickHouse storage, SQL access, and a query API. It does not offer a managed cloud service. ## Product - [Product overview](https://example.com/product): capabilities, architecture, and limitations - [Pricing](https://example.com/pricing): per-seat and per-event tiers ## Docs - [Quickstart](https://example.com/docs/quickstart): Docker Compose to first event in ten minutes - [Query API](https://example.com/docs/api): endpoints, auth, rate limits ## Optional - [Changelog](https://example.com/changelog): release history
What to know before using the draft
- Every link comes from the crawl. The AI selects pages by index and writes descriptions from their titles and metadata, rather than creating new URLs.
- The crawler does not render JavaScript. Websites whose navigation is available only after client-side rendering may produce incomplete results.
- The crawl runs on our servers, and a large language model processes the collected page metadata to draft the file.
- The result is a starting point, not a substitute for editorial review. You should decide which pages best explain your website and what readers should see first.
Deploying the file
- Serve it as plain text at the path it covers: /llms.txt for a whole-site guide, or /docs/llms.txt for a guide to /docs. Check that the URL does not return an HTML fallback page.
- Declare its location so agents do not have to guess. v2 added the
describedbylink relation, which you can publish as<link rel="describedby" href="/llms.txt">in the page head or as an equivalentLink:response header. - Check that every listed link resolves and that robots.txt does not disallow the path. The deployment playbook includes a full verification checklist and a CI example.
- Regenerate the file when navigation or important pages change. An outdated guide can send readers to the wrong content.
Frequently asked questions
What is llms.txt?
It is a plain Markdown file that tells AI systems what a website is and which pages matter most. It sits at the root of the path it covers, whether that is the whole website or one section. The llms.txt entry explains the standard and its current adoption.
Is llms.txt the same as robots.txt?
No. robots.txt describes which crawlers may fetch which paths. llms.txt is a curated guide to the pages an AI system may want to read first. They serve different purposes. See robots.txt and AI crawlers.
Does Google read llms.txt?
Google’s documentation states that its AI features do not use llms.txt. Other major platforms have not clearly documented support. Publishing the file is inexpensive, but it should not be treated as a ranking signal. The llms.txt entry tracks the available evidence.
Do I need an account?
No. You can generate three files per day without signing in. Google sign-in raises the limit to ten per day and stores only your email, name, and avatar. See the privacy policy for details.
Where do I put the generated file?
Place it at the root of the path it covers. A whole-site file goes at /llms.txt, while a file for /docs goes at /docs/llms.txt. When several files apply, agents use the most specific one. The deployment playbook covers serving, verification, and updates.
Why is my result missing pages?
The crawler considers up to 96 pages from the first sitemap files, then selects the pages that best represent the website. llms.txt is meant to be selective, not a complete index. Pages that return 404 during the crawl are removed.
Can I generate one for just a subfolder?
Yes. Enter a URL with a path, such as example.com/docs, and the file will include pages below that path. Under v2, a file covers its own path, and agents use the most specific file when several apply.
Is the generated file spec-compliant?
The structure follows v2 of the llmstxt.org specification: an H1 site name, a blockquote summary, and H2 sections of links. Additional links may appear under Optional. You still need to publish the file, add discovery links, and provide any Markdown versions of your pages.
Keep reading
- llms.txt: the standard, adoption evidence, and limits
- llms.txt deployment playbook: verification and update guidance
- robots.txt: crawler access rules and how they differ from llms.txt
- AI crawlers: categories and access trade-offs
- Free AI Crawler Access Checker: see which AI crawlers can reach your site
- Free Citability Checker: see how easily AI systems can quote a page
- Free Schema Markup Checker: review a page's JSON-LD