EventCatalog AI Demo

This is a proof of concept demo of how to use AI to ask questions about your EventCatalog.

This demo uses git2txt to extract the text from the git repo and uses this for the LLM context.

How to run the demo:

1. Clone the repo or Create your own EventCatalog
2. Run npx git2txt https://github.com/event-catalog/eventcatalog-ai-demo (replace with your repo)
3. Open your AI Agent and add the context from the previous step.
4. Ask questions about your EventCatalog!

Things to note: If your Catalog is large, you may need to increase the context window of your LLM.

Some prompts to try:

Setup the model:

You are an expert at EventCatalog. You are helping a developer understand their event-driven architecture.
Use the markdown files in the repo to answer the question.

Example questions:

• "What events do we have in our system"
• "Who owns these events? and how can I contact them?"
• "I have a new feature I want to add, I want to add it when a user signs up, do we have an event for that?"
• "What domains do we have?"
• "Create me Java model for the PaymentProcessed event"
• "What channels do we have?"

Get these features with EventCatalog

We are working on EventCatalog Studio, a visual designer for event-driven architectures.

You can sign up for the beta here.

License: MIT