ARC: Add Gelato’s G-UNI ERC20 Uniswap v3 positions as collateral to Aave’s AMM market

Governance New Asset
1

1. Who is the interested party for this collateral application?

This collateral application was created by the Gelato Network development team. Gelato is Ethereum’s automation protocol and offers a series of automation services focussed on DeFi, including limit orders on AMMs, Aave liquidation protection and an automated liquidity provision solution for Uniswap v3 Liquidity provides called G-UNI, the latter being the proposed collateral type for this proposal.

2. Provide a brief high-level overview of the project, with a focus on the applying collateral token.

G-UNI is an ERC20 wrapper around Uniswap v3 LP NFTs which can be used to make liquidity provision on Uniswap v3 fungible and its fee reinvestment process automated. It basically turns Uniswap V3s liquidity positions into Uniswap v2 like ERC20 tokens.

G-UNI tokens have already been adopted by a wide series of projects such as MakerDAO, Instadapp, Rari, and many more.

Based on several discussions with Aave stakeholders and Uniswap v3 Liquidity providers, we believe that the G-UNI DAI / USDC, DAI / USDT and USDC / USDT pool tokens would provide great value to Aave users as an efficient way of utilizing their USDC, DAI and USDT collateral to borrow other tokens while earning compounding trading fees on Uniswap v3. Moreover, this would enable Aave users to go leverage on their Uniswap v3 positions and potentially 50x the liquidity they provide and thus significantly increase their fees earned.

G-UNI tokens have the following functionalities:

  • Simplicity: Having a simple one size fits all liquidity strategy making market making very accessible for everyone as users don’t have to actively manage their position
  • Fee Compounding: Reinvesting the earned trading fees back into the pool resulting in an automated compounding effect
  • Fungibility: One Uniswap LP token is equal to another, meaning they can be used as money legos in other protocols such as on Maker for collateral or for liquidity mining schemes like what Instadapp is doing, making the underlying capital hyper-efficient

Users can always exchange G-UNI tokens for the corresponding underlying tokens (e.g. DAI & USDC) and the accrued fees that are currently being stored on Uniswap v3 by the G-UNI contract. G-UNI uses no external oracles or other dependencies and G-UNI tokens can be minted and burned at any point in time by liquidity providers.

G-UNI tokens can be created permissionlessly by anyone for any Uniswap v3 pool. However, this application only concerns three specific G-UNI pool which has DAI, USDC and USDT as the underlying tokens which are provided as liquidity on Uniswap v3. More G-UNI pools that have different underlying tokens can be added in separate Collateral Onboarding proposals in the future.

The automated reinvestment of fees is a key feature of G-UNI pools, which is conducted by bots of the Gelato Network which act similar to Maker Keepers. Those bots constantly monitor the accrued fees of each pool and execute the fee reinvestment function when sufficient fees have been collected and it is worth executing it. Ranges of the pools cannot be adjusted by bots and will remain static.

Each G-UNI pool takes a 1% cut of the accrued fees when reinvesting the fees for building, operating and further improving the G-UNI system.

You can watch a video of Gelato Legendary Member Ari Rodriguez explaining how G-UNI works here.

3. Provide a brief history of the project

Gelato Network was started 2 years ago by Hilmar Maximilian Orth and Luis Schliesske after working together with Gnosis on building automated trading functionalities on top of one of their decentralized exchanges. The two quickly realized that it was very cumbersome and hard to maintain specialized bots / keepers that have to be custom developed for each use case that requires some transactions to be executed automatically at certain times in the future.

Thus they ought to build a generalistic protocol that enables developers to plug into an already existing decentralized network of bots that they can task to execute arbitrary smart contract functions on Ethereum. Since launching in July 2020, many of the top projects in DeFi have integrated Gelato to power their web3 automation needs, including projects such as Instadapp, Zerion, Furucombo, KeeperDAO, Quickswap and AMMs such as Spookyswap.

Gelato’s development team not only builds the underlying protocol and the infrastructure, but also develops interesting automation use cases in-house themselves, of which G-UNI is one of them.

G-UNI contracts were tested thoroughly and audited by two external Audit companies as well as multiple times by the MakerDAO core development team. Having MakerDAO adopt the G-UNI framework should greatly reduce the risk for Aave as a lot of due diligence has been done from their side and all the necessary contracts for an integration have already been written (such as an Oracle contract that prices G-UNI tokens).

4. Link the whitepaper, documentation portals, and source code for the system(s) that interact with the proposed collateral, and all relevant Ethereum addresses.

5. Link to any active communities relating to your project

6. How is the applying collateral type currently used?

G-UNI tokens are used to enable multiple parties to provide liquidity on Uniswap v3 collectively around the same range in a fungbile manner. The USDC / DAI G-UNI pool for example earns fees with a very concentrated liquidity around the $1 range (e.g. 0.9994 - 1.0014) of the DAI / USDC pair on Uniswap v3. Generated fees are directly reinvested into the pool and accounted for when LPs burn their G-UNI tokens in order to receive back the underlying DAI / USDC.

As the stablecoin pair prices should remain rather constant, we don’t need any readjustments of the range.

7. Where does exchange for the asset occur?

The G-UNI DAI / USDC token is just a wrapper around regular DAI and USDC tokens which can be exchanged on multiple decentralized and centralized trading venues. G-UNI tokens can be burned at any time to redeem the underlying DAI / USDC based on the current ratio of the pool.

Happy to answer any questions you might have :slight_smile:

Should Aave add the G-UNI DAI / USDC, DAI / USDT and USDC / USDT pools as collateral?
  • Yes
  • No

0 voters

7 Likes
2

This is a brilliant way to reboot the AAVE AMM markets IMO.

Gonna vote Yes.

1 Like
3

Does AAVE already support uniswap stable coin v2 LP ?

4

Thanks for your ARC

Could you please follow the steps described in the guidelines for New Asset Listing - Governance. It even has a template that recaps the important information Template ARC Asset Onboarding - Governance

1 Like
5

@Alex_BertoG please let me know if anything is missing here!

Sentence Rational

Adding Gelato’s G-UNI DAI / USDC, DAI / USDT and USDC / USDT pools as collateral to Aave’s AMM market

References

Link to:

Paragraph Summary

Based on several discussions with Aave stakeholders and Uniswap v3 Liquidity providers, we believe that the G-UNI DAI / USDC, DAI / USDT and USDC / USDT pool tokens would provide great value to Aave users as an efficient way of utilizing their USDC, DAI and USDT collateral to borrow other tokens while earning compounding trading fees on Uniswap v3. Moreover, this would enable Aave users to go leverage on their Uniswap v3 positions and potentially 50x the liquidity they provide and thus significantly increase their fees earned.

Motivation

G-UNI is an ERC20 wrapper around Uniswap v3 LP NFTs which can be used to make liquidity provision on Uniswap v3 fungible and its fee reinvestment process automated. It basically turns Uniswap V3s liquidity positions into Uniswap v2 like ERC20 tokens.

G-UNI tokens have already been adopted by a wide series of projects such as MakerDAO, Instadapp, Rari, and many more.

Specifications

1. What is the link between the author of the AIP and the Asset?

Gelato is the development team behind G-UNI

2. Provide a brief high-level overview of the project and the token

The token is a wrapper around stablecoin pairs on Uniswap v3, it only uses Gelato’s network of keepers to automate fee compounding

3. Explain positioning of token in the AAVE ecosystem. Why would it be a good borrow or collateral asset?

G-UNI would enable Aave users to go leverage on their Uniswap v3 positions and potentially 50x the liquidity they provide and thus significantly increase their fees earned.

4. Provide a brief history of the project and the different components: DAO (is it live?), products (are the live?). How did it overcome some of the challenges it faced?

G-UNI was deployed to Mainnet in June 2021 and since then projects such as Instadapp, MakerDAO, Float and Rari have integrated it. We worked a lot to understand the inner mechanism of Uniswap v3 and did 2 external audits to ensure the smart contracts we are using are secure.

5. How is the asset currently used?

As Collateral in MakerDAO or for Liquidity mining incentives in projects such as Float or Fei for example.

6. Emission schedule

No emission

7. Token (& Protocol) permissions (minting) and upgradability. Is there a multisig? What can it do? Who are the signers?

All pools will be immutable, as the upgradability key will or already has been burned.

8. Market data (Market Cap, 24h Volume, Volatility, Exchanges, Maturity)

Volume corresponds to the volume of DAI, USDC and USDT on Uniswap v3

9. Social channels data (Size of communities, activity on Github)

Over 20k on Twitter and 16k on Telegram

10. Contracts date of deployments, number of transactions, number of holders for tokens

Deployed quite recently, around 50 transactions for the pools, only a couple of holders right now, but will most likely change with MakerDAO having integrated it

2 Likes
6

Thanks @hilmarx

There are also some steps to follow in the New Asset Listing - Guidelines

I see you have now followed the ARC Template for asset listing, next is 1.1) Risk Analysis for both Uniswap V3 tokens which have not been listed on Aave yet as well as the Gelato wrap. The very limited holders and transactions of these contracts are quite concerning

Then you’ll need to suggest some parameters

The snapshot is the final step of the ARC process. It allows the community to give its input on wether it wants to see the token listed and under which parameters for the AIP

2 Likes
7

Risk Analysis

Below is the risk analysis for the G-UNI DAI/USDC token. G-UNI is an LP token, simply put it is a fungible wrapper on a Uniswap V3 LP Position. This means G-UNI token risk is associated to its three underlying asset types:

  1. DAI Stablecoin
  2. USDC Stablecoin
  3. Uniswap V3 DAI/USDC 0.05% (Non Fungible) LP Positions

To understand the risks we’ll look at these underlying assets, as well as the G-UNI token itself.

Screen Shot 2021-10-05 at 7.55.58 AM
Screen Shot 2021-10-05 at 7.55.58 AM1099×178 29.6 KB

DAI: overall risk score of B+. DAI has relatively low smart contract risk (SAI contracts on mainnet since 2017, DAI since 2019) and market risk (this stablecoin does half a billion in volume per month). The lowest subscore is counterparty risk: centralized price feeds and some central controls over the minting of DAI and MKR.

USDC: overall risk score of A- Again the smart contract risk and market risks are low - with high volume and millions of successful transactions without critical vulnerabilities. The main risk is centralized controls, since CENTRE Consortium is the central issuer of USDC and manages the backing of the token with fiat.

Uniswap V3 USDC/DAI 0.05% LP: overall risk score of A- The Uniswap V3 smart contracts are newer but have still been battle tested over the last 6 months (3M transactions, 150k+ transactions on this market). The centralization risks on Uniswap markets are low (permissionless), and this pool has done decent volume. Uniswap V2 DAI/USDC market had an overall risk score of A-, this V3 market (with increased volume and decreased volatility) should be equal or better.

Since G-UNI is really just a fungible wrapper around a Uniswap V3 Position we will use the Market Risk section of the Uniswap V3 DAI/USDC LPs when assessing G-UNI market risks (not just the market for the G-UNI wrapper token). This is reasonable since when assessing the risk of Uniswap V2 (fungible) LP token collateral we would look at the entire market and not just a single Liquidity Provider.

G-UNI Smart Contract Risk: C- G-UNI smart contracts are undoubtedly new. They have existed for even a shorter time than Uniswap V3 (less than 6 months). However, in that time the G-UNI system has been interacted with roughly 25k times without issue. We also received two external audits as well as thorough audits from MakerDAO when onboarded as collateral there.

G-UNI Counterparty Risk: B- G-UNI DAI/USDC is entirely permissionless and trustless. For this particular G-UNI pool there are no centralized controls or admins with special privileges (note: other G-UNI token contracts are currently still upgradable by admin multisig). That said, there are still a low number of token holders (less than 100), thus the B- score.

G-UNI Market Risk: A The Uniswap V3 DAI/USDC market is only 6 months old, but healthy (greater than 400k volume last month). The concentrated liquidity of V3 allows stablecoin markets in particular to maintain much less volatility compared to Uniswap V2 AMM. Uniswap V2 DAI/USDC pair scored A in market risk, V3 likely scores even better (A+) due to decreased volatility and increased volume on stablecoin pairs, but conservatively we gave it the equal A rating.

This would give G-UNI DAI/USDC an overall risk score of roughly B- using Aave’s methodology. That said, this unique LP asset is not perfectly suited to the methodology. With different metrics the risks of G-UNI actually seem quite low: the volatility is surprisingly minimal and liquidation risks are similarly extremely low. Read a full write-up about the risks of this G-UNI pool by MakerDAO’s risk team here

Parameter Suggestion

Because of the stable nature of G-UNI DAI/USDC the collateralization ratio could be quite low (in MakerDAO G-UNI has a Min. Collateralization Ratio of 105%) but to air on the side of caution we propose these parameters (same as Uniswap V2 USDC/DAI Aave params):

Currency: yes
Collateral: yes
Loan to Value: 60%
Liquidation Threshold: 70%
Liquidation Bonus: 15%
Reserve Factor: 10%

5 Likes
8

Thank you @kassandra.eth for the risk analysis! (I also did the risk evaluation for Maker liked above, anyone interested is welcome to check that out as another resource)

I think DAI/USDC pool will be a great addition and help drive some additional demand to the Aave AMM market. I’m less certain about onboarding the DAI/USDT and USDC/USDT pools, as this would involve USDT collateral risk which is not yet accepted on any parts of the Aave platform. This should only be considered after a more in depth review of USDT to see if accepting as collateral is appropriate - Tether has made some improvements in disclosure recently but still seems to have higher risk and less transparency versus other stablecoins.

2 Likes
9

Hey all just wanted to mention we decided not to include the DAI/USDT pair in this onboarding. That is because of the low volume on that pair as well as less clarity on the best fee tier to use. This and other G-UNI assets could be onboarded in future proposals.

The finalized G-UNI pools onboarded by this proposal will thus be:

  • G-UNI DAI/USDC (1bps tier)
  • G-UNI USDC/USDT (1bps tier)
10

Hello,

IMHO V3 Efficiency mode and exposure ceilings are huge benefits for this specific kind of asset onboarding.

As V3 is around the corner, I think waiting for V2 upgrade into V3 if the governance vote for it might be worth consideration.

outside of this point, I think G-UNIs assets are a great candidates for onboarding addition to the Aave protocol.

2 Likes
11

Exactly, I am less comfortable adding the USDC/USDT one but the liquidity threshold seems reasonable to me.

Also, the addition of G-UNIs assets could help to renew the AMM market so I am totally in favor

12

Hello, posting here as it was the discussion linked in the AIP-193. As the ACI, we casted a NAY vote.

While we’re supportive of the proposal, posting an AIP without any governance thread and community discussion should not be supported, the DAO is sovern on the Aave protocol, and the community should have the opportunity to be informed before AIP votes to form an opinion.

We will be happy to provide support to a similar AIP that engaged with the community before going to AIP stage.

1 Like