# llms.txt generator & checker > Create an llms.txt file for your website from your sitemap, with page titles filled in, or check an existing llms.txt against the llmstxt.org format. Free. Canonical: https://cromagnon.app/tools/llms-txt-generator Last updated: 2026-10-27 - Category: SEO - Formats: LLMS.TXT ยท MARKDOWN - Use it: https://cromagnon.app/tools/llms-txt-generator (free, runs in the browser) ## What it does llms.txt is a plain Markdown file at the top of a website, at example.com/llms.txt, that gives AI assistants a short, curated guide to the site: what it is, the key facts, and links to the pages that matter most, each with a one-line note. It was proposed in 2024 at llmstxt.org as a companion to robots.txt and sitemaps, written for language models rather than crawlers. This tool builds the file for you, starting from your sitemap if you like, and checks existing files against the format. ## How to use it 1. Type your domain under Start from your sitemap and click Load pages, or type sections by hand. The tool finds the sitemap through robots.txt and groups pages by folder. 2. Delete pages that don't help someone understand your site, such as tag archives and old posts, and move secondary pages into a section called Optional. 3. Click Fill in page titles to read each page's title and description (up to 40 pages), then edit the notes so each says what the page offers. 4. Write the name, a one- or two-sentence summary and any details an assistant should know, such as regions, prices or what you don't do. 5. Read the checks, download llms.txt and upload it to the top folder of your site. 6. To check a file, switch to Check llms.txt and type the site or file address, or paste the file. Tip: Write the summary and notes as you would brief a new colleague: plain facts, no slogans. An assistant quoting your llms.txt should give an accurate answer about what you do. ## The format - **# Name**: The only required part: one top-level heading with the name of the site or project. - **> Summary**: A blockquote right after the heading with a short description and the key information needed to understand the rest. - **Details**: Optional paragraphs or lists with more context, such as how to read the site, important limits or terminology. - **## Sections**: Headings for groups of links, each followed by a Markdown list: - [Title](https://...): optional note. - **## Optional**: A section with this exact name holds links that can be skipped when an assistant needs a shorter context. ## Does llms.txt help with AI search? Be realistic. llms.txt is a proposal, not a standard, and as of late 2026 the major AI search products haven't said they use it for ranking or citing. Google has said it doesn't support it in Search. What it does reliably is help when someone, or an agent, points an assistant at your site: coding assistants and agents read llms.txt files, many developer documentation sites publish them, and a clear file gives any model that fetches it a fast, accurate summary instead of a guess from your navigation. It takes about half an hour to make and costs nothing to host, so for documentation, software, e-commerce and services sites it's a sensible low-effort addition. It doesn't replace good pages, a sitemap, structured data or robots.txt rules about AI crawlers. ## Building one from your sitemap Load pages reads robots.txt for Sitemap: lines (or tries /sitemap.xml), follows a sitemap index to a few of its sitemaps, and takes up to 150 pages, skipping images, files and addresses with query strings. Pages are grouped by their first folder: /docs/ pages under Docs, /blog/ under Blog, and top-level pages under Main pages. The home page's title and description are used for the name and summary if those are empty. That's a starting point, not a finished file. An llms.txt is most useful when it's short: the 20 to 50 pages that explain your product, prices, policies and key guides. Delete the rest or move them to Optional. Fill in page titles opens each page once and uses its title and meta description; rewrite the notes so they describe content rather than market it. ## What the checker looks for It confirms the file starts with a single # heading, has a > summary, uses ## sections with list items in the - [Title](URL): note form, and has no deeper headings, which the format doesn't use. It flags list items that aren't links, relative links, duplicates, empty sections, very long files and a file that isn't at the top of the site. If the address returns an HTML page instead of the file, which happens when a site sends its home page for unknown addresses, it tells you that too. Edit this file loads a checked file into the generator so you can fix it. ## Tips - **Keep it current**: Regenerate the file when you add products, change prices or retire pages; stale facts are worse than none. - **Plain text**: Serve it as text/plain or text/markdown, not as an HTML page, and don't block it in robots.txt. - **Link to Markdown too**: Documentation sites often add .md versions of pages and link those, which are cleaner for models to read. - **Consider llms-full.txt**: Some sites also publish llms-full.txt with the full text of key pages in one file. - **Facts first**: Put opening hours, regions, prices and contact details in the summary or details, where they're found quickly. - **Be honest**: Describe what you do and don't do. Assistants may repeat it to people deciding whether to buy. ## Questions ### Where does llms.txt go? In the top folder of your site, so it opens at https://your-domain/llms.txt, like robots.txt. ### Is llms.txt the same as robots.txt? No. robots.txt tells crawlers what they may fetch; llms.txt gives AI tools a summary and a reading list. To block or allow AI crawlers, use robots.txt. ### Is llms.txt an official standard? No, it's a community proposal from llmstxt.org. Support varies by tool. ### How long should it be? Short: a name, a summary and a few dozen links is typical. Long lists belong in your sitemap. ### Can I add it on WordPress or Shopify? Yes. Upload it to the site's root folder on WordPress, or use a plugin; on platforms without file access, some SEO apps can serve it. ### Is anything stored? No. The file is built in your browser. Sitemaps and pages are fetched once by our server for the import and aren't kept. ## Related - [robots.txt Generator](https://cromagnon.app/tools/robots-generator) - [Sitemap Generator](https://cromagnon.app/tools/sitemap-generator) - [Schema Generator](https://cromagnon.app/tools/schema-generator) - [Markdown Preview](https://cromagnon.app/tools/markdown-preview) - [All tools](https://cromagnon.app/tools)