Felix Leupold

One man's “protection” is another man's “patronization”

CowSwap is now powering @aave. Swap collateral or debt or repay with collateral, all in a seamless experience and best execution with MEV protection.

Few people seem to know that there is a @safe module allowing to guard transactions from being executed before a certain timestamp: github.com/schmanu/multis… H/t @schmanu_

@koeppelmann @hal2001 I think the goal of ve-tokens is to create friction (not all token holders may be able to stake) and thus boost value accrual for a fraction of token holders. This breaks Modigliani–Miller. Also they allow to boost long term commitments (locking) which burn programs don't.

@hal2001 and if someone really wants a token that pays dividend - you can actually engineer that on-top: x.com/koeppelmann/st…

Worth correcting: Yes, share buybacks are in fact economically equivalent to paying dividends. UNI buybacks are economically equivalent to paying dividends to UNI holders (but easier to implement and more tax efficient). This was proven in 1961 by Modigliani–Miller.

This could work for ERC777 tokens (which call back the sender on transfer, so it could adjust their order) in combination with smart accounts. I'm not sure there are a lot of users/tokens with this setup though. Otherwise I'm afraid this would be purely off-chain semantics, which could easily overstrain the size of the bond solvers are putting up with in case the limit orders are very large.

@CoWSwap can we get another order type? we have fill-or-kill, we have partial fills, what about 'until balance depletes'? If I place a limit order for 1m USDC to 1m USDT, can you reduce the fillable amount by any USDC that moves out of my wallet (transfer, swap, anything) after that point? e.g. place limit order at 12:00 for 1m USDC to USDT 12:04 I give up on waiting and swap 500k elsewhere instantly, balance remaining 500k usdc 12:00 next day, I transfer more USDC into the wallet now the order's max fillable amount is 500k because order was for 1m at 12:00 but you can see outflows of 500k since then = 500k remaining is this technically possible? would save so many ppl so much in laptop repairs @fleupold_

resting stale limit orders on cowswap are so fucking annoying holy shit

It’s one of those mornings where you wake up, see there was a crash, and check how much extra $ you made from your @CoWSwap limit entry bids getting executed better than their limit price 🤌 Only place where a “limit order” can mean more than you asked for.

« You cannot create a backdoor that only lets the “good guys” in. » Let alone tell who the good guys are these days.

@haydenzadams @0xKakumei +0.25% UI fee

@0xKakumei Like less than 0.1% lol

I was at an airport in Europe recently and spreads on euro / usd were 20% This is the most liquidity forex pair on earth The absolute greed of tradfi is unmatched and defi will replace it

@PossibltyResult Isn’t utilisation at like 85% so the extra 4% would only accrue to the 15% of USDC/T that’s not borrowed?

The first DeFi protocols to find ways to pass stablecoin yields through to depositors will have a major advantage DeFi yields are risky (smart contract risk, financial risk, custodial risk...), meaning they have a high hurdle rate to overcome and attract depositors An extra 4.2% would basically double USDC and USDT yields on AAVE (with >$12B in size)

@ameensol I don’t know, but I think PBS showed that taking choice away from individual validator/builders and “protocolizing it” works as a defense to do otherwise legally questionable things. Builders don’t manipulate the market, they just include bundles given to by “a protocol”.

The case against FOCIL (which I learned about today): ETH devs, I love you. You mean well. But when you create an EIP to solve the problem of "filtering out transactions with sanctioned addresses" and your solution is "to allow validators to impose constraints on builders by force-including transactions in their blocks"... we have a problem. A big problem. And if you don't see it, you're either being naive or reckless. So far, the censorship resistance story on Ethereum has gone something like this: 1) ETH staking is permissionless to join 2) stakers can decide to include txns... so even if 99% of staking nodes censored tornado cash txns for example, it would just take 100x longer This has worked out, in my opinion, more or less OK. Even at peak Tornado Cash censorship, only 90% of nodes were censoring TC txns, so txns should have taken 150s instead of 15s. This setup allowed node operators in the US who could potentially face legal risk for interacting with and facilitating txns to the sanctioned addresses (up to 20 years in prison for a sanctions violation) to simply filter them out and continue participating in the Ethereum network. Even with the block builder oligopoly, only 2 out of 3 of the block builders are censoring, and as the OP mentioned, 90% of the rest of the validator set is NOT engaging in censorship. What FOCIL does is FORCE INCLUDE TRANSACTIONS FROM SANCTIONED ADDRESSES, such that VALIDATORS CAN NO LONGER CHOOSE TO FILTER THEM OUT. This is potentially a big problem for US validators, who now could face legal penalties for staking. It appears the "plan" is to attempt to limit the legal liability of the validator chosen for each block by distributing the responsibility of deciding on txn inclusion across a set of attesters (other validators who are not chosen for the block) such that each block's validator can claim "well look, I didn't choose the txns in the blocks, I'm just validating them." I don't have great faith in this plan. The US government could: 1. Decide it doesn't care about the whole "attester" business and go after known validators who include sanctioned address txns in their blocks anyway. 2. Decide to go after the known attesters who decided to include a sanctioned address txn in the block. 3. Go after the core devs who designed a system to coerce the validators into including sanctioned address txns. If you don't think they won't do #3, that's weird, because I didn't see you at either Alexey's or Roman's trial. If I was the US gov, I would actually be 100% in favor of FOCIL. You mean to tell me the ETH validators are all going to be *forced* to incriminate themselves by validating blocks with sanctioned address txns? Well great, that means I can go after any ETH validator on US soil whenever I want, seize all their ETH, and prosecute them for a sanctions violation. And you're telling me the ETH core devs made their *intent* to coerce validators into violating sanctions publicly available on the ETH research forum? Amazing, now I can even round up any relevant ETH core devs that might pass through the US and prosecute them for conspiracy as well. It also doesn't help that, while FOCIL is currently not designed to provide any incentives validators to include specific txns, instead relying on "altruism" (here, altruism means the willingness to include txns from sanctioned addresses, bc of course it does 🤦‍♀️) this is also not the plan forever. Currently, the plan is Option 1 - to simply rely on "altruistic" validators to attest to the sanctioned address txns, and not provide incentives to the attesters. But I don't think the plan is for things to stay this way forever, as there is an active research effort into figuring out the best way to reward the attesters for their service. The paper below was published May 2025, and includes Julian Ma, one of the original authors of the OP. In conclusion: This isn't 2019 anymore. We can't afford to be naive about the implications of the systems we design and build anymore. FOCIL would coerce ETH validators to include sanctioned address txns, and in doing so, could call into question the legality of ETH staking generally. FOCIL could even incur legal liability on the core devs who designed and implemented it, because it was *explicitly* designed to stop validators from filtering out sanctioned address txns. I love you all, but let's not. cc @soispoke @barnabemonnot @fradamt @_julianma

It's to explain MEV protection in an intent based solver competition and the price advantages that not crossing the spread when settling a multi dimensional batch auction with uniform price clearing brings to the average crypto user in one word.

@0xcuriousapple @0xpessimist This (programmatic orders for EOAs) is on our roadmap. Until then people can use a Safe and the StopLoss App there. We were hoping smart accounts would take off in the meantime, but as long as people keep using EOAs every dApp has to write their own intermediate contract.

@0xpessimist yeah they have programmatic orders framework, which could be used for something like this i dont get why cowswap hasn't done that yet tho why no stoploss for normal EOAs with an intermediate contract in between used for approvals and execution ? @fleupold_

It’s 2025, and why there isn’t any dex offering stoploss orders

@fiddyresearch It's relevant because it says the value of a company is independent of the way it's financed. Doing a buyback + issuance is equivalent to paying a dividend + issuance or simply keeping the revenue in the first place. I wonder what form of financing you think is more effective?

@fleupold_ No clue what that is or why it is even relevant tho. Today’s probably not like how tradfi used to be 75 years ago.

whenever I see a good protocol I respect talk about doing buybacks ...

Great work by the whole team - from research through development to making a smooth transition in production! @CoWSwap just got the biggest mechanism upgrade since inception and is now faster and fairer for all.

Probably the most ambitious project from @GnosisDAO yet. If money is just social coordination, why not use your actual social circle to create your own IOUs and use them as money. Will probably take some more time for people to get, but imo a very interesting experiment...

Paradigm preparing to become a solver…

@mo_nokh @CoWSwap While we cannot change existing AMM behavior, we can work towards making more and more people part of the CoW Protocol batch. Obviously, we need to ensure fairness for all participants, ie. they get at least the same outcome as they would in a non batched world.

In my vision every token should have a single price per block no matter if traded by an investor a passive liquidity provider a liquidation or else. Failure to do so is precisely what causes MEVs in various forms (🥪, LVR, OEV).

The relationship between users, solvers and CowAMM has become quite precarious. If the goal is to maximise user output, can any shifting be considered not excessive? 🧵