We have designed a "reproducible paper" template[1] for  managing our research outputs by studying best practices, including  RDA recommendations and outputs, since 2015[2]. It has since significantly grown, and is now a fully documented template.

We already use it heavily to start new projects, and to propagate fixes and new features to ongoing projects. It is usable on a personal computer, or a large server; automatically downloads input data and necessary software (by PID where available); builds and installs the software; does batch analysis by running the software on inputs to  produce outputs in a highly closed environment; finally, it finishes  with a dynamic PDF paper/report[4] (where figures and numbers will  be updated if the relevant part of the analysis is updated). After publication, the output datasets, the customized template, and its version controlled history, are publicly released with PIDs.