Me too, I thought whoever found the private key would go through Mara's slipstream, that no one would dare go through mempool directly but how wrong I was.
The thief who stole the coins must have a nice setup, it took his bot 45 seconds to replace the transaction, I assume whether he used BSGS or Kangaroo, it took 30 seconds to find the private key and the remaining 15 seconds for opening and shutting down apps.
kTimesG bragged about having a very fast setup and also started precalculating tames for 66bit, 67bit... Preparations you would only need if you plan to steal the transaction sitting in the mempool.
Every other user that took part in the stealing-bot-script discussions a few pages ago is suspicious.
The thief likely active here but would never confess because of repercussions...and well he would unmask as an asshole

.
I'm tending to think it wasn't any of the usuals in this thread 'cause we all thought the solver would go through MARA. For his very reason I didn't even bother to finish developing mine.
Also, I don't think anyone would bother to precompute against 66 in this scenario as the key can be cracked in mere seconds anyways.
If that was the case it would have taken much less than 45s to replace the original tx, as once you have the key it might take less than 1 second to relay the first replacement.
I was also thinking.... What if the creator himself triggered the attempted withdraw?
Reason: To officially embed this event in the challenge's timeline at the cost of 6.6 BTC, as now "puzzle #66" will be notoriously known to have been hijacked by a bot in the proposed scenario.
Am I making any sense here or am I just hitting my bong too much?
