Also, the exploit detailed to Wired was reported to Secret, fixed by the startup, and then they broke the fix, allowing the vulnerability to be exploited again.
But at some point in the last few weeks, as the company expanded its infrastructure, the bot detection system somehow failed, allowing Caudill and Seely to rediscover the hack, Byttow says.
That's the kind of shit that QA is supposed to track and fix. They probably used an old version of code to do their expansion, and that's just as sloppy as digested chili.