Swell recently published a research report by Gauntlet that aims to create a framework for selecting Actively Validated Services (AVS’s) to inform the design of rswETH.
To explore the report, we were joined in a Space by Carson from Gauntlet, Ismael from LAGRANGE and Jon from HyperLane.
In case you missed it, here are the top questions from the event.
Why is our recent AVS Selection Framework report important and how is it relevant right now?
“Right now, the payment slashing conditions from EigenLayer have not been published, and so a lot of frameworks have lagged behind waiting for that information to be available. From conversations with the Swell team, it became obvious that some of these frameworks needed to be established now, and so we started developing those frameworks and mathematical processes for maximizing risk adjusted yield. What we did was take an approach of some assumptions around flow of funds, payments and slashing, and put together this framework for how an AVS might approach AVS selection. I think it does a great job of introducing some of the inputs into this optimization function that LRTs are going to need to consider. As payments and slashing conditions become clearer, these functions and equations will become more concrete, but with that being said it's nonetheless important to get these things established now especially as AVSs begin to build out. This is so they can understand how LRTs are approaching the problem and enable better progress for other builders in the space, given the limited information we have at our fingertips right now.”
– Carson, Gauntlet
In terms of risk and reward, how is restaking to secure AVS’s different from staking to secure Ethereum?
“With restaking, you are reusing that stake to validate other services beyond what you see right now. Within the LRT space we are seeing different risk profiles across these different AVS’s, and across those you will also see different payment terms as well. From an LRT perspective. what you’re thinking is “Do I want to opt in to validate this service as well for the potential increase yield while I take on additional slashing risk into my LRT?”.
– Carson, Gauntlet
“Every AVS is going to have to denominate the rewards to staking and capital providers either in their native token, USDC or ETH. The question of how much fees to pay is predicated on the risk profile that the underlying AVS presents to its users. In terms of designing slashing conditions, I believe there are some basic principles that all teams should be following to minimize the risk of some slashing event occurring to their capital providers and operators. These basic ideas will be things like ensuring that you have the conditions set that you use for slashing so you don't have these subjective conditions that could potentially be incongruent with having a safe AVS, or ensuring the slashing conditions you have are clear, transparent and very easy to avoid.”
– Ismael, LAGRANGE
“It is ultimately about how you price risk. The common view with the restaking market is comparing it to securitization, but I think it's much closer to an insurance market where we think about how we are going to pay out the AVS’s or what the AVS lets us do that we couldn't do before. Usually that is something that is translated to an end customer. We want to use this to bring rollups canonic security, so they aren't just relying on the validator committees they are using today. We think that this will unlock more activity for those rollups, and that is something worth paying for.”
– Jon, Hyperlane
How does your protocol think about risks and rewards of restaking with AVS’s?
“With LRTs, one of the major things we’re looking for is what are the operators that we work with best suited to run in terms of AVS’s.This is why Swell has been talking extensively with operators and AVS’s as this is a part of the market that is quite new, which is why we are moving as slow as we can, and making sure we are covering every aspect of risk on the operational side.”
– Abi, Swell
“It is important to think through how much security an application needs. There is a large amount of LRT capital available right now, and a protocol should be deliberate in how much pooled security and attributable security it needs to underpin this design.”
– Ismael, LAGRANGE
What are your thoughts on how we can begin to categorize AVS’s?
“I think you can go incredibly deep in terms of categorisation. There are the obvious points such as yield, slashing risk and correlation which will be important. Then you have the standard categorisation such as interoperability, zk specific bridging, data availability, and then other things such as, “Does the AVS have an insurance fund to cover the risks of slashing?”.
– Carson, Gauntlet
“Gauntlet’s framework doesn’t just capture what the AVS does, but it captures who is involved in both the operation and capital providing for that AVS. Those are two of the most important heuristics of AVS’s as the ecosystem matures, because you have the imperative on a number of AVS designs to have globally distributed operators and varying capital sources.
If you want to ensure an effective and safe underlying network that doesn’t have single points of failure at the operator set, or the set of capital providers, then I think that beyond that the characterisation of what the AVS does is important as well. That is very important as a lot of AVS’s will tend to stack risk if you have someone validating for both of them, and the thing that they are validating for is very similar. This can be prevented once you have effective modelling.``
– Ismael, LAGRANGE
“The exciting thing with LRTs is the amount of diversity, categorisation and risk profiles available actually creates potential for a decent amount of diversity across the space. Whereas in LSTs the risk profile stays pretty much the same.”
– Carson, Gauntlet
“There is going to be a very different expectation on yield for attributable security versus pooled security. These are two very different types of stake, and I think we are going to see the rates people are going to have to pay for these stakes are very different. If multiple slashing positions trigger, then in theory you will still get back whatever the attributable security you paid for is, which is a higher rate. If you just have the pooled security, the amount you get back will be less.”
– Carson, Gauntlet
Thanks to everyone who attended the event!
To stay up-to-date with Swell, join our thriving community on Discord, Farcaster, DeBank, and X.
