Hello! I have been working on kubeconform for quite a while, a re-implementation of kubeval that solves a few of the issues with kubeval that bothered me most:

• Some warnings cause kubeval to exit with 0, some with 1
• The output is not ideal - -o json will output non JSON on top of the JSON output, same with -o tap.
• the complexity of working with CRDs, though the (undocumented :() -additional-schema-locations is pretty close to what I added.
• the use of comma separate values for -d is not very practical, can't use wildcards for example
  Kubeconform also downloads/validates over multiple go routines, so might be a tad faster.

Reasons for making a reimplementation, rather than sending patches to kubeval - was that some of the issues required significant re-architecting to implement - which makes for very large patches. Without acccess to the kubeval repository, it is very difficult to iterate efficiently.

Kubeconform has mostly reached feature-parity. The CLI, the API, and the outputs are all slightly different though.

This ticket is to track interest in merging both projects, the work that would be required to do so, and potentially discuss the level of compatibility we should aim for. I expect this might take a little bit of effort.

• Is there interest / time to merge both projects?
• Should I build an alternative CLI that is 100% compatible with kubeval? I do find the cli of kubeconform easier to work with and the use of goflags makes it easier to test. However 100% compatibility might make migration easier?
• Should I build an API wrapper to support kubeval's API in kubeconform?

I would happily participate in the maintenance of the resulting project.