Each replication attempt lives in its on github repository within the replisims organization.
I suggest using the structure commonly used for R packages with an adaptation for research compendia as reccomended by Marwick et al.:
Marwick, B., Boettiger, C., & Mullen, L. (2018). Packaging Data Analytical Work Reproducibly Using R (and Friends). The American Statistician, 72(1), 80–88. https://doi.org/10.1080/00031305.2017.1375986
This can easily be implemented using the rrtools` package.
You don’t have to use git I would higly recommend it though as it eases collaboration. See the resources section for some recommendations on how to get started.