Skip to main content

LLMS.txt

View as Markdown

Enable tools like Claude, ChatGPT, GitHub Copilot, and Cursor to quickly understand your EventCatalog.

What is LLMS.txt?

LLMS.txt is a proposed standard that helps AI-powered development tools better understand and interact with your documentation. Similar to how robots.txt guides web crawlers, LLMS.txt provides structured information that makes it easier for AI assistants like Claude, ChatGPT, and GitHub Copilot to process your EventCatalog documentation.

The file is automatically generated and maintained as part of your documentation pipeline, requiring no manual configuration. It organizes your documentation's key concepts, structures, and relationships in a format optimized for machine reading.

llms.txt and llms-full.txt

The llms.txt file includes your EventCatalog resources in a simple format, listing each resource with a short summary.

The llms-full.txt file includes your EventCatalog resources in a more detailed format — all the contents of your catalog resources are included in the file.

Both files cover events, commands, queries, services, domains, flows, channels, teams, users, entities, data products, and ubiquitous languages.

Enable in EventCatalog

llms.txt is enabled by default in EventCatalog.

You can disable it by turning it off in your eventcatalog.config.js file.

eventcatalog.config.js
llmsTxt: {
enabled: false,
},

Access the files

The recommended URLs for AI tools are at the catalog root:

The legacy paths /docs/llm/llms.txt and /docs/llm/llms-full.txt still work for backward compatibility.

How to use LLMS.txt?

Once you deploy your EventCatalog you can use your tools to ask questions about your Catalog.

LLMS.txt