AI Safety Research

It's just responsible disclosure, the sandboxing package repository likely was deployed elsewhere and needs fixing before the details get published and exploits get tried out.

No doubt they've borrowed here and there, but you seem psychotic and Hanlon's Razor seems more fitting here.
( https://umbrex.com/resources/tools-for-thinking/what-is-hanlons-razor/ )

There is a specially good chance now when even https://ai-2040.com/?choices=plan-a-root authors advocate "total AI research transparency"!