For example, when Project Orion was being seriously considered the scientists had to find ways of making large quantities of fairly powerful nuclear weapons cheaply and quickly. Based on something Freeman Dyson said, I think they succeeded to some extent, but that secret now has died with the scientists who worked on it.
There has to be a lot of writing squirrelled away somewhere, because there are restrictions like "You agree to obtain a validated export license when exporting if this product is incorporated into the design, development, production, or other activities related to chemical weapons, biological weapons, nuclear weapons, or ballistic missiles." but no available literature on how these packages may actually be used in this context. (https://welsim.com/download)
One of the reasons Nuclear Testing is now very uncommon is because computers and software are now advanced enough to simulate them accurately. And yet, despite that, there is no "Nuclear Weapons design: A modern approach" available for public consumption. These are worked on by physicists so someone must be wasting time by writing books somewhere.
The moral implications of such a device are fraught. To use it is to detonate the very weapons that one should not detonate.