I'm working on a state resolution algorithm for #Atoki today (which, who knows, might make it into #Matrix some day too). And I could use some help.
Please send me your most cursed event graphs that you would expect to break a state resolution algorithm and be exploitable! I'd like to check that it can deal with these cases *before* deploying it out in the wild.
Don't really care about the format, as long as it describes the graph well enough to understand the weirdness. Diagrams are fine, DOT syntax is fine too, textual descriptions too.
@arianvp The problem with fuzzing stateres is that the determination of "is this outcome reasonable" is very difficult to make if you do not already have a known-good stateres mechanism, because you're testing against user expectations and not against a singular technical property.
@arianvp (And well, we don't have any known-good stateres mechanism, that I know of...)
@joepie91 have you considered fuzzing?