Ordinarily, documentation is either hand-written or generated from code. This turns source code into the single source of truth for a program and documentation is always a second-class citizen. In the literate programming paradigm, the code and documentation are on equal footing, since both are generated (documentation is said to be woven, source code tangled) from the same source files.
I maintain my Emacs configuration and website as literate programs in Org files. On this page you can find the documentation that is woven from these Org files. Although the code could be tangled from these files, I include the code in the Git repository for two reasons:
To avoid the code going out of date with my Org files though, I add a step that checks the tangled code matches precisely what has been checked into Git.
Note that all of these are still in their very early stages, but here you can find documentation for the following programs: