I noticed recently that OpenAI had added JSON to the Playground dropdown options for Code sharing (they probably did this a while ago, I just noticed now) so this seems like a good time to re-up the standard that I created and am using to simplify keeping track of my presets. As I say in the README,
This data model provides a standard you can adopt to separate content from logic in your applications, to make it easier to construct complex prompts, and to make it easier to share detailed presets with other developers.
This is open source and public on Github so I would be very interested in thoughts and suggestions from the many knowledgeable folks here.
If there is interest, I have also created a parser that reads the PresetJSON into a pandas dataframe, supplies sensible defaults if they are missing, and passes the information along to the OpenAI completion endpoint.