Generate a compliant llms.txt file for your site so LLMs index the right pages. Spec-compliant.
Free llms.txt Generator builds a compliant llms.txt file for your site — the new standard (adopted by 600+ sites including Stripe, Anthropic, and Hugging Face) that tells language models which pages are worth indexing. Fill in site name, tagline, description, and one or more sections with URL lists — get a spec-compliant llms.txt file ready to drop at your domain root.
Privacy: This tool runs entirely in your browser. Your text is never uploaded, logged, or cached. Close the tab and it's gone. Verify in DevTools → Network: zero requests fire.
Every site should ship an llms.txt so LLMs surface the right content for user queries. Making that free-and-easy is part of keeping the open web healthy. The generator runs 100% in your browser.
| Input | Output | Typical size | Speed | Login needed |
|---|---|---|---|---|
| Form fields | llms.txt (markdown-formatted) | Any site size | Real-time | No |
Here's what this free tool does in detail — every feature is built to solve real problems, runs entirely in your browser, and is free forever.
Follows the spec at llmstxt.org exactly — H1 site name, optional blockquote tagline, optional description paragraph, H2 sections with bullet lists of links, and an optional Optional section at the bottom for secondary content. Every link is a markdown bullet - [Title](url): description.
If you want to understand the emitted markdown, render it with the Free Markdown to HTML tool or preview with the Free Markdown Live Preview.
llms.txt files typically group related pages into sections like Docs, Guides, API Reference, and Blog. This generator lets you add or remove sections on the fly, each with its own title, optional description, and URL list. The URL list accepts one link per line with optional title and description separated by em dashes or pipes.
To build a related AGENTS.md file for your repo, use the Free AGENTS.md Generator.
The spec recommends an Optional section at the bottom for links that are useful but not core — changelogs, community pages, release notes, archives. Toggle it on and paste secondary URLs there; LLMs know to treat these as lower-priority context. Toggle off if you're keeping things minimal.
Switch between Full URLs (https://example.com/docs) and path-only URLs (/docs) depending on whether you're deploying to a single domain or building a portable file. The Extra spacing toggle inserts blank lines between sections for a more readable file.
Once you have llms.txt, pair it with a good README — the Free README Generator builds one in minutes.
Site name becomes the H1; tagline becomes a blockquote; description becomes an intro paragraph.
Each section gets an H2 title, optional description, and a URL list. Add as many as you need.
Save as llms.txt and drop it at the root of your domain. Also consider an llms-full.txt with expanded content.
Ship an llms.txt so ChatGPT, Claude, and Perplexity pick the right pages when users ask about your product.
Make your docs site discoverable so LLMs answer support questions from the docs, not guesses.
Curate the paths that matter — Pricing, Docs, Changelog — so LLMs don't surface random blog posts.
Build an llms.txt as part of your dev-rel site launch checklist.
Keep the LLM-facing index synced with the site's actual structure.
Yes. Free forever, no account required.
A markdown-formatted index at the root of a site that tells LLMs which pages to surface. Adopted by 600+ sites including Stripe, Anthropic, and Hugging Face.
At /llms.txt on your domain. Some sites also publish /llms-full.txt with expanded content.
One URL per line. Optionally add a title and description with em dashes or pipes: 'https://example.com — Title — Description'.
No. Runs 100% in your browser.