One of the core conceits of ForensicDbg was that if we build a significantly improved crash debugging core, that value would carry through, regardless of the final workflow.

Our testers have helped us confirm that on the more traditional, visual debugging side. But how do you confirm that for AI debugging workflows?

With a crash dump evaluation gauntlet, of course!

If you just want the results, or even grab a pdf with the raw testing results, that is all at the bottom.