Platform engineering is the discipline of designing and building toolchains and workflows that enable self-service capabilities for software engineering organizations in the cloud-native era.

Platform Engineering

apparently not, there are open requests for this feature. For now it has to be manual steps /  `run` implementations

I'm kinda confused by the question. You should be using a golden pipeline to push images into a repo and then using said repo. I might be misunderstanding, but this sounds like you haven't decoupled the image building from its use...

Or maybe you're asking a question about layer caching in Github... their implementation seems unnecessarily complicated, but very doable.

<@U04RM6TSJ84> yes it's not about using the image, it's still steps before releasing it.

what you say also works but I'd avoid pushing an image immediately after build. First I want to test &amp; secure it and then push it.

But the build action cannot simply `push` it has to build &amp; push because it requires `context` and `path` (Dockerfile). So it would end up being:

• build image (locally)
• test
• security scan
• build again and push
I want to avoid build again

That may be the source of confusion for me. I'm not building things locally, I let the CI/CD workflows do that. Some testing happens before its pushed to the repo, more testing happens after. Verified builds get tagged. Builds that don't pass get pruned.

<@U04RM6TSJ84> sorry I am making you more confused haha

I am not building locally, it's in CI. By locally I mean that the output is an image or tar file, in the Runner (no local machine).

I said locally because buildkit argument type for this is `local`

Ok... cool... that's better. :sweat_smile: