I'm the author. I'm seeking HN feedback for how to improve this CVD documentation, because AI attacks are escalating and some are turning up new kinds of security issues.