I went through this thought process with SIT[1] -- it started out with issue tracking hard-coded into it but it has been later extracted into a separate model and the data & process models are being refined to make them much more generic and support other use cases.

[1] https://sit.fyi