This is a project meant for static analysis of vmp3 and has been forked from git.back.engineering although mine has the advantage of actually working. Well, except for the VTIL compiler that is. The vtil compiler was ripped from https://github.com/vtil-project/VTIL-Utils and it is a broken demo that needs fixing. Also, there are multiple vmenters per actual routine and most don't do that much so it would also be nice to stop treating vmenters as functions, instead inserting an unconditional jmp in place of a vmenter to the compiled vtil and then an unconditional jump back so IDA can display it better. Mainly focused on semantic accuracy of the vtil compiler though at the moment (and not crashing which it likes to do)
Update: This project has been abandoned for a while due to updates to VMProtect and also VTIL turned out to be inadequate and I will need to learn LLVM or something similar to achieve what I initially set out to do. Anyway, I will definitely return to static analysis of VM-obfuscated binaries in the future, because I think that obfuscated software is the antithesis of digital freedom and safety from cyber-threats, it is everything I dislike about proprietary software dialed up to 100. There needs to be a free and open source toolkit for reverse engineering VM-obfuscated binaries.