What's the point of bluefin-cli if we are working inside containers?

[luiz-brandao]

Hello, I just stared using Bluefin. I installed bluefin-cli and it looks really cool, that said I think I'm missing a piece of the puzzle. If everything is immutable and I'm supposed to work with containers, what's the point of having bluefin-cli if whenever I enter a container to do some work I loose all the goodies? For instance, I'm trying to setup a Java dev environment and to build the project, run CLI tools, I need to enter de container, but once I enter I have no more goodies! So either I will have to install the tools inside the toolbox or run commands with a prefix... Help me understand. Thank you!

[ehsanullahjan]

for cli tools, bluefin generally recommends homebrew. you can use brew install ... to install cli tools (e.g., zoxide, bat, etc.). since multiple java projects might require different jdk versions, you can manage per-project jdks and related toos/envs via mise (brew install mise).

alternatively, you can use devcontainers for fully isolated, per-project dev environments.

finally, you can also create either per-project distroboxes or one global distrobox that serves as your primary dev environment (latter is more practical). inside your distrobox, you'd have to install tools you want to get the kind of ux you're looking for (e.g., starship, zoxide, bat, lazygit, etc.) -- this directly answers your original question. again, for per-project env/sdk setup and auto activation, you'd want mise installed in your distrobox.

bluefin gives you many options you can choose from based on your taste, and workflow requirements. however, unless you have something against it, homebrew is the easiest approach that just works.