On this page we go over all of the concepts to be aware of. The AI model is aware of these and will make sense of it from the requirements provided. There's no specific way to articulate any of these concepts, simply express them in the way that suits your writing style: it's all language. The only recommendation is that for best results you will want to be as clear as possible. In general, if something might be confusing or unclear for a human, it will be for the model.
Entities represent collections of records similar to sheets in an Excel workbook or tables in a database. Entities have fields/columns that store and present the various data points. Entities often have relationships to other entities and the model will extract those relationships. Entities and their relationships serve as an important foundation for information that can be extracted later for reports, query fields, dashboard widgets and more.
There are a few default entities that are always present to support system functionality. The main one to be aware of is for system users, which serves as the collection of user records that can sign into the application with fields like first/last name, email, password. Additional fields can be added to these records. In the event the field you're adding already exists, you'll see a warning to let you know.
There is also the concept of initial values or entries for collections. This comes in handy if you have a collection that supports the ability for users to add to it in the future but you have an initial list to populate. For example, if you needed to add a list of car brands, language along the lines of "Add initial entries for Ford, Toyota, Tesla..." would be acceptable.
Below are the various field types that can be hinted at using any language desired (e.g. Field A as a datetime field). Whenever a type isn't specified or inferred it will default to a text field.
Below are the various settings you can include as part of your requirements for fields to provide a better data entry experience as well as higher standards for data.