if i can offer my opinion, i believe i see only one solution to the concerns raised by the comments preceding mine: there is a rhyme, in a conceptual sense rather than phonetic, to the one we’ve developed here in this playground of cents on the dollar. github dot com/QuidLabs/IMO/blob/main/src/MOulinette.sol#L444
you may observe how creating an LP deposit is done in the most efficient way possible. if there’s a source of the necessary second token already on our books, then we can simply move this from one form of being accounted to another, no slippage, no questions asked, just adjust. only do the actual swap (incurring slippage) if absolutely necessary.
this is USDC, though, and we have a particular incentive model for how we attract USDC specifically. now that we’re armed with some perspective, back to the original issue in question: we need ETH to clamp onto FOLD and deposit into Uni. where are we going to get it? well…there’s an AVS sidecar. that means ETH that’s being re-staked is passing through it.
here’s what is possible. instead of re-staking the ETH directly. you can let it sit, and grab it on-demand whenever someone wants to lock FOLD into Uni. in the exact amount that you need to satisfy that particular LP deposit at that particular moment in time.
doesn’t that get in the way of all the AVS footwork? interestingly, not! we can use the LP deposit as what gets re-staked. Eigen has this cool feature that they shipped in August that allows you to use any ERC20 for re-staking.
now…the only question is…how the heck do we get an ERC20 out of the LP deposit. this is solved in a very particular way with quid. it will not apply directly to the problem we’re trying to solve together here. with that said though, i believe it shines some light on an approach. and we can figure out the details of that approach if this sounds promising.
Bunni has been explored, but in my personal opinion, it’s not the simplest solution and we might come up with something even better.