LLMs.txt generation

An llms.txt file is a machine-readable document that helps AI crawlers discover, understand, and cite your website's content. Think of it as a sitemap.xml designed specifically for large language models — it tells AI engines what your site contains and where to find it.

Vizzybl generates these files automatically by crawling your website and producing AI-optimized descriptions for each page.

llms.txt vs. llms-full.txt

Vizzybl generates two complementary file types.

File	Purpose	Best for
llms.txt	A compact routing file listing your site's major sections with short descriptions	Directing AI engines to your most important pages quickly
llms-full.txt	A comprehensive file including all discovered pages with detailed descriptions	Maximum content discovery across your entire site

Most sites benefit from both. The llms.txt file acts as a directory, while llms-full.txt provides full coverage.

Generating your files

1. Open the Create section and click the LLMs.txt button
2. Enter your website URL (e.g., example.com)
3. Choose how many pages to crawl under Pages to crawl (20 by default)
4. Click Generate

Every run starts a background job that produces both llms.txt and llms-full.txt. Progress is tracked in real time, and you can close the tab and return later without losing progress. Smaller runs finish sooner.

How many pages a run can crawl depends on your plan:

Plan	Pages crawled per run
Free	20
Pro	200
Ultra	500
Enterprise	1,000

The page shows your plan's limit before you start. Choices above it are shown but locked, with a link to upgrade. On a site with more pages than you chose, Vizzybl crawls up to that number and builds both files from those pages.

Documentation sites

To build the files for your documentation rather than your whole website, tick This is a documentation site and enter the address of your documentation, such as example.com/docs or docs.example.com.

• The crawl stays inside that address: pages elsewhere on the site, such as pricing or the blog, are left out.
• A documentation sitemap at that address (e.g., example.com/docs/sitemap.xml) is used first when there is one.
• The files are titled as documentation, for example "Example Documentation", unless the site's own name already says so.

Reviewing and refining

After generation, Vizzybl displays a split-screen workspace:

• Left panel — An AI chat interface for refining your files
• Right panel — An editor showing the generated content with tabs for llms.txt and llms-full.txt

Use the chat panel to refine your files with natural language. For example:

• "Rewrite the descriptions to emphasize our enterprise features"
• "Group the URLs by product category"
• "Add schema type mentions for each page"
• "Remove all blog post URLs and focus on product pages"

Each refinement updates the editor in real time.

Exporting your files

Once you're satisfied with the output:

• Copy to clipboard — Click the copy button to grab the content for a single file
• Download as ZIP — Export both llms.txt and llms-full.txt together

Upload the files to your website's root directory (e.g., example.com/llms.txt). AI crawlers look for these files at the root path, similar to robots.txt.

Tip: Regenerate your llms.txt files periodically — especially after adding new pages, products, or content sections — so AI crawlers always have an up-to-date view of your site.

Next steps

• Citation optimization — Optimize your page content to earn more AI citations
• Content rewriting — Rewrite content for higher nugget density and readability
• Brand protection — Monitor how AI engines represent your brand and close vulnerabilities
• Setting up your brand — Configure your brand for accurate tracking