Roman Agureev

Предлагаю переименовать DeFi в GdeFi — Gde мои деньги, Лебовски?

@real_philogy Have you heard of Vyper?

After 6 months of work, we're proud to finally share our first release of our new smart contract language: Plank v0.1 🚀 To fix the fundamental issues plaguing smart contract development we're rebuilding the language stack from the ground up. 🏗️ Learn more 👇

@newmichwill add_liquidity_i_love_kim_jong_un

Finally figured out how to protect all my smart contracts from DRPK

pov you're layerzero:

@duonine @Morpho This is the risk you take in return for yield, otherwise all rates should be equal

Imagine this. USR peg drops to 20 cents, buy USR. Use 1 USR to borrow 1 USDC on @Morpho Peg is hardcoded to 1. Easy 80% profit. Repeat until no more USDC left. LP drained. Thanks for playing crypto.

@elonmusk Only way to check is to overpopulate the Earth and see if we start glitching

I had dinner once with a top physicist and a top computer scientist and asked what they thought the probability was that we were in a simulation. They answered simultaneously at 0% and 100% respectively. It was like a double-slit experiment, but with humans.

@newmichwill New security era is “if you’re AI…” comments

The end of security by obscurity is near!

@StaniKulechov You’re not fully right on “the only protocol” with such protection. Morpho has optional bad debt socialization too, though not split between 2 kind of deposits

It’s been a remarkably resilient week for the Aave Protocol and DeFi as a whole. Over the past seven days alone, the protocol liquidated more than $450M worth of collateral across multiple networks. For a $50B+ lending protocol, this represents roughly 0.9% of total deposits at the time. At the same time, Aave has continued to accumulate additional revenue, helping offset risk while rewarding participants. What’s truly fascinating is the scale Aave has reached. No other system is able to manage resiliency at this level today. That said, lending is never a free lunch. Bad debt can occur as part of the process, and Aave has built-in mechanisms designed to handle exactly that. The system is backed by overcollateralized, liquid assets with pre-vetted risk parameters, a critical factor in any lending context. Aave’s resiliency doesn’t rely solely on its technical design, but also on the rigorous risk frameworks established by its risk service providers. This combination of engineering and risk discipline is what sets Aave apart. Pre-vetting risk parameters, performing accurate due diligence, and maintaining robust modeling are key reasons Aave has earned trust over the past five years. This resiliency is further reinforced by a broad, autonomous liquidation network that actively competes for liquidation opportunities, ensuring efficiency even during periods of stress. On rare occasions when excess debt is not fully covered by liquidations, it can be automatically absorbed by Umbrella pools, Aave’s native protection mechanism. Aave remains the only protocol with dedicated, built-in user protection of this kind that users can partake. Learn more about Umbrella here: aave.com/docs/aave-v3/u… All of this together defines Aave’s resiliency. The protocol has weathered multiple market cycles, from the FTX collapse and centralized lender failures to Black Monday, Thursday, and Friday. The takeaway is clear. DeFi wins through resiliency, transparency, and a superior cost structure. We’re also releasing a new liquidation engine for Aave V4, bringing even more flexibility and performance to the protocol. More details here: aave.com/blog/aave-v4-l… If Aave and DeFi can withstand the most volatile asset class in the world, crypto, we should apply these systems to all assets, across both crypto and TradFi. Aave will win.

@DefiMoon Are you somehow sponsored to fud curve? If YB LPs begin to withdraw arbitrage cycle will take place and bring the peg back…

🚨 $crvUSD depeging again as borrow rates have exploded to 40% crvUSD/USDC 0.985: If bitcoin starts cascading here due to $MSTR crashing, crvUSD will depeg very hard as the pegkeepers are empty and the pools are very imbalanced. There's only $2m of exit liquidity in each of the top crvUSD pools on Curve, so it won't take much to crush the crvUSD peg right now. The ratio is 10:1 😬😱 If @yieldbasis LPs start to panic at some point and begin to withdraw their BTC, the downward pressure on the peg will increase substantially!! Like I said many times before, CurveDAO was irresponsible to approve not only the initial 60m crvUSD credit line, but also the subsequent 300m credit line for YieldBasis without extensive real-world testing. RIP $YB $CRV $CVX 🪦

@nobrainflip Russian prostitutes

Gold is crashing Silver is crashing USD is crashing BTC is crashing ETH is crashing Stock market is crashing Real estate is crashing If everything is crashing, where the fuck is money even going?

@FraxForce crvUSD PegKeeper debt ceiling was increased to 9M iirc

@TokenBrice Despite not relying on centralized assets, a stablecoin still needs liquidity for on/off-ramps. New PegKeeper candidate for @fraxfinance on @CurveFinance?

Founder Mode: activated Objective: deliver the stablecoins DeFi deserves Status: pkill stealth.bash && ./load wp.tex ▓▓▓▓▓▓▓▓░░░░░░░░░░░░ 42% tokenbrice.xyz/why-polaris/

@austingriffith It has nothing to do with decimals, it's yield part of tokens

⚖️ Balancer Hack TL;DR: 🧮 Most tokens on Ethereum use 18 decimals, but some don’t. ✖️ Balancer upscales token amounts (to 18-dec) and downscales them back. 🔧 Upscaling always rounded DOWN, but downscaling could round UP or DOWN. 😬 The more scaling steps involved, the greater the chance tiny amounts are lost in edge cases. 🎛️ An attacker tuned swaps to hit those rounding edges. ♾️ Each hit gave a tiny profit, so he repeated it hundreds of times in a loop.

@fiddyresearch @MarcoWorms Rounding occurs at any div operation. It's usually easy to see what the error is wrt divisions (be it constant or variable), nonetheless the final formula can be bulky. Didn't find any such tool for python to port to vyper. I see it as a tool which highlights errors for each var

my initial thought here is that this is more of a dev-culture thing rather than some tangible solution. for e.g. in vyper, all external calls are explicitly stated (you need to do `extcall blabla(params)` so auditors know what's up. soon, you need to decorate your functions with a reentrancy decorator to explicitly mention that reentrancy is occurring. maybe whenever rounding occurs, there needs to be a GIANT red flag somewhere that immediately fails the smell test of auditors and reviewers.

my take here is we 'solved' reentrancy by fire. and now we need to solve rounding.

@Zodomo @fiddyresearch Balancer did use FixedPoint library

@fiddyresearch Rounding is simple. Accrue that 1 wei to the protocol, rather than the user. Always round down on the user. They're not going to miss that 1 wei. Solady FixedPointMathLib already fixes this.

@k06a @DefimonAlerts I don't think so, every round amounts are ~10 times smaller meaning it's a loop to drain say 9/10 each round

@DefimonAlerts It seems that Balancer exploiter inflated their LP token share with series of weird swaps of LP=>token0 and LP=>token1, and then token0<=>token1. Potential rounding issue of ComposableStablePool. dune.com/maxmorrow/tx?t…

Новая ОСь apple — это downgrade. Первое впечатление от 17-го айфона в руках — китайская подделка.

@fiddyresearch Freedom is another abstract idea to hide intention under a nice “modern” wrap.

Remember this, Freedom is a pure idea. It occurs spontaneously and without instruction.

@0xDanchou @fiddyresearch @crosscurvefi This will actually help crosscurve as arbitrage cycle will be much faster

@fiddyresearch does this make @crosscurvefi useless

Curve FastBridge. An idea theorised almost a year and change ago. Now brought to life by Roman (curve core dev) with a significantly simplified design: just use LayerZero + don’t fuss too much about L2<>L2 transfers (yet?). The gist is simple: we don’t like so many versions of the same token so consider the L2 main bridge minted token as canonical, use a message passing infra to notify layer-1 vault that a deposit occurred, and transfer ‘unbacked’ token (debt) from layer-1 vault to use. The token that the user bridged back to L1 (that is stuck in the challenge period) ‘backs’ the L1 debt that was transferred to the user. This works only for stablecoins that you can mint unbacked, and does quite good things for L2 ‘peg’. Bridge can be changed by LZ does a pretty good job. If tether likes it, so do we. github.com/curvefi/fast-b… Proposal to offer a credit line to FastBridge protocol: gov.curve.finance/t/fastbridge-f…