Gloated Rollups and AVS Compression

[sambacha]

AVS Compression

Note: This is 1/2 of the 3rd post, the other half, Gloated Rollups will be out before Friday.

Aggregate all use cases with an operator/delegation based slashing condition

Restaking benefits are not strictly limited to having a ‘traditional’ AVS, meaning as a separate service running alongside a validator client (e.g. MEV Boost like sidecar). You could be staking simultaneously to insure a lending protocol (e.g. AAVE), alongside supporting the more usual examples of oracle/keeper networks.

This is where the idea of AVS Compression comes into play.

N of M problem for AVS

While no two AVS services are the same, validation services can fall into broad categories, which we will call service models. A service model might target simple validation services like a basic price oracle: a setup with a data feed, consensus mechanism, and state management. The service model just needs to know which validation logic to run, how many operators to include in the set, and the data verification rules. The logic of the service model then expands that service configuration.

Figure 0: N of M AVS Problem

For example, the service model would transform the list of operators into a configuration for the restaking infrastructure, with specific collateral requirements and reward distributions (i.e. it determines the eligible enrolled validators for rewards and for slashing conditions out of the subset of validators).

Yes it really do be like dis

Covenant Enforcement

To this end, we created a continuous enforcement system, this was born out of testing for XGA among all the validator clients to ensure conformance, this is from back in June during the inital Lido testing phase:

Figure 1: NUT, Network under Test^[1]

So instead of having to adapt to the idiosyncrasies of each infrastructure provider, service models can target a standardized representation (i.e. the ‘covenant’)

Multi-Restaking AVS

If you’re dealing with multiple AVS service models and multiple restaking protocols, you have to maintain integrations between N service models and M restaking providers. This integration includes both generating each protocol’s specific configuration and also deploying it— a process which can vary significantly across different restaking implementations and their underlying validation requirements.

Figure 2: N of M reduced by Covenant

Validation Conformance and AVS Management via Covenant

• Instead of managing each AVS separately, operators could define their intended state:

  • Total stake allocated
  • Performance parameters
  • Risk tolerances
  • Resource allocation
  • Validation rules

• Safety Mechanisms:

  • Implement dependency checks between protocols
  • Ensure changes don’t violate risk parameters

We can reduce, this to the following validation formalism:

\forall \text {, input } \in SM,: \operatorname{valid}(\text { input }) \Rightarrow \underset{SM}{\operatorname{valid}}(\text { output }) \text { where output }=\underset{SM}{\operatorname{request}(\text { input })}

The Juice

Two key opportunities exist:

• Insurance Funds: Leveraging locked assets for underwriting (i.e. $FOLD via Captive Insurance)
• Options and Derivatives: Providing collateral for margin trading. (i.e $.LST’s)

Captive Insurance

The Captive Insurance basically idenmifies validators from slashing conditions less a deductible. It is underwritten by staked $FOLD. $FOLD in turn earns a fee from the services it covers, which could potentially be more than just our AVS service. By extending the coverage to additional AVS’s, we in turn become a meta-AVS, which enables AVS Compression bringing higher returns than if an operator selected the same set of AVS.

Margin Trading Funding

This is where XGA benefits in that traders wanting to leverage their position can do so with collateral rehypothecation, earning fees by lending to the XGA Auction participants.

Bonus: Restaking Derivatives

• Restake the derivatives into new protocols.
  • Accidentally invent DeFi 3.0 chaos.

Forecasting

Manifold Boost Forecasted APY. Note, that Today does not include XGA at all, only the improvements made on top of MEV Boost + Validator setup that are guaranteed to produce the additional yield. The source for this google sheet is from Finality Capital’s Restaking report, which is provided below:

Manifold Boost Google Spread Sheet https://docs.google.com/spreadsheets/d/e/2PACX-1vRXiLLIH_WRUA3m7zckL8yJXuwclX1fBabsvS2fMuxRXdaL755gJUy_BcnwccCmfrnmTqXDK1bJzyzA/pubhtml

Original Google Spread Sheet https://docs.google.com/spreadsheets/d/e/2PACX-1vRa3bqGuewG870DQ0Eq2w7Cr3FDumowP3zuDGW0NK5KBDvTlTDtCpyLEFKqWt2crZ5tvy_E2_Qe30xK/pubhtml

Finality Capital Partners ‘Restaking Report’ PDF https://www.restakingreport.com/

---

---

1. In remberance of Peanut. ↩︎