Out of the 3, 2 are a pointless waste of tokens. 'create-XXX' and 'deploy-XXX' are repeatable, reproducible, scriptable tasks that were solved automation problems (not necessarily easy, but solved) long time before LLMs. If you're going to use a Claude for anything like that, it's much better to ask claude to actually write a deployment script or a template for new projects that can run in a deterministic, tested way. Using an AI skill for a process like that is slower than a script, less trustworthy and just a waste of money every time it runs.
I get it that some people don't know how to write shell scripts or IaC templates, so in theory maintaining a LLM skill is easier than maintaining a makefile, but if you have a coding agent available, it can do the updates.