Thanks for the review @Stefano Bonicatti and @alessandrogario on 5616. There's great notes as well as some considerable follow up recommendations. I do not have a lot of time to dedicate to this right now. My goal with the PR was to unblock the stagnant pre-built deps (both the deps themselves and the model for containing their logic). I do not have time to make the directional changes you suggest. Maybe the best option is to close the PR and hopefully use some of the changes as inspiration for a more sustainable dependency/prebuilt solution, and at the very least the build logic ported from Homebrew/Choco to CMake?