Regarding the shared codebase. We developed a general schema for the metadata of journal articles that includes CRediT and other CROs. It is a JSON schema so you should be able to use it and validate against it in your own application. Using a common general schema to store author metadata has the advantage of interoperability. So ideally people should be able to send their information from tenzing to your app and vice versa without any problems. The schema is called belayer and you can find it here with an accompanying example: https://github.com/tenzing-contrib/belayer There is another group that is working on creating a schema similar to this called JAMS but they do not have a working prototype yet (to my knowledge) but once they come up with their own standard we want to accommodate belayer to their specifications (https://github.com/jam-schema/jams).
https://jats4r.org/credit-taxonomy
This is nice to have and low priority.
We could also implement the tenzing scheme:
From Marton Kovacs: