As the author of asciinema I'm flattered that Rodrigo considers asciinema as a potentially useful tool for this problem.

This solution would likely apply to a small subset of contributors, unless most Dillo hackers are terminal editor users, and would probably help only partially (which he acknowledges in the post). I guess partially is still better than not at all. Setting aside the technical aspect, I wonder how practical it would be to review long asciinema sessions when reviewing a patch.

If you're reading this Rodrigo: is your recording verification idea to just quickly eyeball / skim through a recording to catch human behavior in those sessions? Or maybe something more elaborate?

Btw, I'm not taking sides here, just happy to see asciinema in the wild :)