Showcase: Use GPT-4o vision to add events to your Calendar from screenshots

I built a cool iOS Shortcut that takes a screenshot of your phone, use GPT-4o Structured Output to extract events, and add them to Calendar.

You can read the full story here: Reddit - Dive into anything

I see that you send the base64 encoded image to openai as text, how does it work for you? Does it extract the text correctly from base64?