Actually the allowed character count was not enough, so I had to truncate my original question. My full question is: How come there is no successful (not already abandoned) FOSS alternative to Ansible written in a typed language ?

My main pain points regarding Ansible: - It becomes really slow above a certain playbook size (I am aware that execution time depends on a lot of other variables) - Has a lot of dependencies (I am looking at you pip & python) which can be painful to manage - no out-of-the-box support for ssh-less/passwordless login (SSM, Hashicorp Boundary, etc...)

I feel that these problems would be more or less solved by a more efficient implementation (distributed as a binary). Of course in order to gain real momentum it would be beneficial for this tool to have a configuration which is mostly compatible with current Ansible roles & templates. The chosen templating language could be jinja2 or any other similar DSL. Given the compatibility, migration would be less painful.

The main points in favour of Ansible and other current configuration management tools: - community - quality/quantity of available plugins and packages - expandability - already established trust (especially important in enterprise) - democratized access enabled by easy to learn script languages (python for Ansible, ruby for Chef)

The new tool would have to cross off each items from the above list in order to be successful, yet I don't feel any of them even remotely impossible.

The question still stands: why?