Compound

Simple summary

This proposal improves the execution and yield of the Compound Aera Vault pilot:

1. Allows the vault to use Milkman from CoWSwap when trading BAT. a. Liquidity conditions have severely deteriorated onchain for BAT since the initiation of this pilot, requiring additional exchange integration with Milkman in order to tap into offchain liquidity
2. Adds cUSDCv3 as an allowed asset in the asset registry, enabling the vault to earn yield and provide liquidity to Compound with its USDC
3. Changes the guardian address, a hygiene update for the Guardian as Gauntlet migrates some internal Guardian infrastructure

This proposal does not change the goals of the Aera pilot (diversification out of BAT and ZRX, and managing the proceeds in a volatility-targeted vault), it is simply an update to the existing vault to better enable these goals. It similarly does not change any of the security properties of Aera, the vault is still non-custodial and only the Compound DAO can withdraw/pause the vault.

Motivation Milkman After starting the pilot, BAT has been very difficult to sell on-chain. So far the vault has been able to sell roughly $55k of BAT and there is another $57k remaining in the vault.

Since the pilot program began, we have added support in Aera for CoWSwap’s Milkman contract with the goal of increasing the available liquidity for BAT swaps. This integration allows us to dynamically adjust execution price as the market price changes and taps into a network of solvers that are able to fill these orders efficiently and competitively using offchain liquidity.

We expect that this will enable the Aera Vault to swap 10x more BAT at same 100bps slippage bounds (increasing daily volume from $300-$800/day to $2k-$5k/day). Of course, this is just based on current liquidity conditions and existing solvers, and given that market liquidity is dynamic this could change.

cUSDCv3 We have also heard the community’s feedback on including cUSDCv3 as an asset in the vault and allocating the existing USDC in the vault into cUSDCv3. The benefits for the DAO would be as such:

1. Provide additional USDC liquidity to the Compound III ETH comet
2. Generate yield on an otherwise idle asset in the vault

In the limited scope of the pilot, the actual need for this capital to cover insolvencies is dwarfed by the other reserves in the system (both V2 and V3 have over >$50M in reserves compared to the <$500k USDC in the pilot).

Full forum post

Specification

This proposal will:

1. Add new assets:
   1. COMP, which we’ll receive as rewards from lending USDC
   2. cUSDCv3 Pricer, which prices the vault’s lending position
2. Accept ownership of new hooks module at address 0x7A8E7F9490fb927d20835b97578E521c97720f64. We need to deploy a new hooks module and have the Governor Bravo timelock accept ownership of it to get around a limitation in Governor Bravo that only 10 actions can be called in a single proposal. Without this workaround, adding the following function sighashes would put this proposal over that limit. This new hooks module already has all the existing hooks, plus the following new sighashes:
   1. Milkman Router: requestSell, cancelSell, claim
   2. BAT: transfer, to move the token into the Milkman swap contract
   3. ZRX: transfer, to move the token into the Milkman swap contract if needed in the future
   4. cUSDCv3: supply & withdraw
   5. Comet Rewards: claim
   6. COMP: approve & transfer, to move the token into the Milkman swap contract if needed in the future
3. Change the guardian to our new guardian address: 0x7587D73Ec14157440c01485555F81F31cF6c549e. (The fee recipient will also change, but this is moot because fees are fixed at 0 for this pilot)