As Gareth noted years ago the CHT has some significant limitations to its customization/functionality. Without a plugin architecture, developers seeking to make deep customizations to the CHT can only fork the repo and modify the core source code. This works, but is a lot of effort for the partners to maintain.
A better approach would be to support a plugin architecture with clearly defined features/interfaces. This would reduce the effort for creating custom experiences in the CHT and make it much easier/safer to maintain those experiences when upgrading to new versions of the CHT.
@Anro has recently re-started the conversation in a serious way both in this GitHub issue and by providing a working Proof of Concept PR! Building on this momentum, I think the time has come to form a formal squad to pursue this feature! (See the docs for details about squads.)
The term “plugin” is very broad (see the examples in the link above). As a squad, we will have to narrow the scope down to a specific MVP that we can pursue first. This MVP should be as tightly scoped as possible, but should still address a real-world need. (Thanks again to @Anro for already providing some initial ideas/examples of both in his PR.)
Please reply here on this thread if you are interested in participating in the squad!