For the remaining unsolved lower bitlength puzzles the solver has to use non-public mempool with so far the only easy public service at slipstream.mara.com OR the solver is a complete idiot as was the case with puzzle #69.

Any solver of puzzles #71, #72, ... which are not multiples of 5 have to be mined from non-public mempools. Period!

Solvers of remaining puzzles which are multiples of 5, i.e. #135, #140, ... #160, don't necessarily need non-public mempools and should be fine to broadcast their puzzle withdrawal transaction publicly. Though it doesn't really hurt to do it via slipstream.mara.com, too. I'd consider it safer that way.

It's funny how some dudes here don't seem to understand how RBF works and that basically all mining pools have an economic incentive to have FullRBF enabled by default. Therefore you can't prevent that your transaction gets replaced as long as the replacer follows RBF rules. You can't opt-out RBF successfully anymore. Period!

Anybody who doesn't understand this, shouldn't play with solving #71+, just don't come later and cry you've been robbed by bots. DYOR!