Nekojita: SushiDAO Treasury Evolution

SushiSwap was launched as an innovation in stakeholder rewards, built on the elegance of the UniswapV2 AMM, but has now expanded to various other DeFi products, including the MISO token launchpad, Kashi lending, BentoBox vault infra, and soon, the Trident AMM system and Shoyu NFT marketplace.

All of these products provide value to Sushi stakeholders, through the xSushi staking and SushiMaker fee router contracts. Sushi has also grown to support 13 different chains in its short life, leading the charge for multi-chain DeFi. Suffice to say, there is a lot happening on Sushi these days.

To date, Sushi token holders have coordinated to support Sushi and launch the above products by snapshot voting and informal governance mechanisms, such as twitter polls to elect representatives in a “Treasury multisig” (with ~$130 million in holdings), which helps manage a “core” Ops multisig and independent dev teams.

While the SushiToken was launched by Chef Nomi with logic for on-chain governance, it was not completely implemented. Thus, Sushi governance has proceeded through game theory around token price and cooperation of the above actors with informal checks-and-balances.

It seems helpful, more than a year later, to examine a few things:

  • Is the Treasury multi-sig still desirable, functioning similar to a Board?
  • If so, should key signers be expected to run for reelection?
  • If not, should the Treasury move to a different smart contract, perhaps a DAO?

This post is meant to broach the subject and gauge the feelings of the Sushi community on these matters.


Going to have to do some reading on this subject to see how the Treasury multi-sig is operating and has operated. If it is operating similar to a Board, I imagine that would be a positive step. But not certain if I think a board is decentralized enough in nature. That is a very centralized governance structure.

Running for re-election would be interesting and informative for those of us who maybe came to Sushi later and are unfamiliar with those on the multi-sig and there roles and responsibilities. Running for re-election would make sense if this is the choice for governance.

DAO operated Sushi seems like the inviting governance contract to me.

I think OPS multisig provides enough flexibility, it is and should be used for daily actions and fundings needs. Making the treasury multisig less needed and more open to on-chain governance.

If we go for classic on-chain governance, I would like to see a multisig of delegates on top.
This multisig could bypass the timelock (for example if default timelock is 7 days, multisig could reduce it to 48 hours) in case we need to execute a proposal quickly (ex: compound fiasco).
They could also put a proposal on hold in case it doesn’t respect the sushi ethos.
When I say “put on hold” it could be like asking for a new vote on a proposal, up to 3 votes and if all 3 votes comes with the same result then it’s executed. Or just increasing the timelock on the proposal execution (ex: from 7 days to 14 days).

I would like to see the community run new elections to choose these guardians.

It might be a “boomer” thing but it could be interesting to try to write down what the Sushi ethos is and put it on-chain, kinda like the “Sushi constitution”.

Looking forward to seeing everyone’s ideas and comments.


Constitution or mission, vision, values are always good ideas. Honestly, think they are different phrases for a very similar thing.


I feel like this is, perhaps, two different discussions in one - or at least could be couched in 2 different timeframes:

  1. the realpolitik of how Sushi multisig/treasury currently operates, i.e; who - why - how were those positions decided - is there a need to reevaluate the position/powers of the multisig holders (2-6 months)

  2. the vision for how Sushi should operate, i.e; what are we aiming for - why - how do we get there (6 months - 2 years)

Given that, I think it’s fair to say that the answer to 2 would be that Sushi from it’s inception was always meant to be moving towards full on-chain, token holder governance of the protocol (imo - I’m open to all discussion of what “on-chain governance” could/should actually look like). Sushi was created and markets itself as a DAO, not a corp with some preference signalling tech. I think starting this discussion up again is well needed right now and could be a catalyst for regaining some of the earlier levels of community involvement Sushi/these forums used to receive.
In saying that, that’s probably many moons worth of discussion and iteration - esp. around aligning the lofty goals of decentralization with the realities of operating Sushi, practically, according to regulatory and market pressures, yadda yadda yadda.

In regards 1) I vibe with @hhk’s idea, but I don’t know if that would end up having the opposite effect by ‘legitimizing’ multisig holders as ‘decision makers’ - rather than automatons who must enact the will of the governance vote. I’d propose reducing the multisig to a “Veto option” and transitioning towards a system where the multisig are considered more a last line of defence against the community voting for something that could completely ruin Sushi irrevocably.

Like learning to swim. At some point you gotta jump in the deep end, but it’d be nice to have some adult supervision for a while.

Random shower thought: On-chain governance is really just da biggest multisig.


What a kickass quote. This is how I feel about decentralization and the trend in that direction.

Fully automated DAO’s are great in theory but hard to achieve. I believe as we transition in this direction is important to have some sort of oversight.

For a multi-sig is there a similar “golden mean” you hope to approach for amount of signers?

A board feels archaic and corporate to me, straying away from the original purpose of DeFi.

How can we achieve this oversight in an efficient way? Would a separate contract for a Treasury DAO save on gas costs and expedite the voting process? If not, I don’t quite see the purpose.

Could we envision a future where someone sourced from the community is a key signer?

1 Like

Given the tools that Sushi already uses, I’d like to propose a potential solution using the recently announced Zodiac standard by Gnosis Guild.

Sushi currently has this forum, a Snapshot space, several Gnosis Safe multi-sigs deployed on multiple networks (and presumably some other governance tools on networks where the Gnosis Safe is not yet deployed).

Given that, I think the natural choice for how to give token holders more legitimate control over Sushi would be to make use of the SafeSnap plugin to Snapshot. In short, this enables a Snapshot space to autonomously control any number of Gnosis Safes on any number of networks, without the need for input from the multisig signers.

I’d propose reducing the multisig to a “Veto option”… Like learning to swim. At some point you gotta jump in the deep end, but it’d be nice to have some adult supervision for a while.

SafeSnap can control a Safe in parallel with the multisig signers, so this type of veto power works out-of-the-box. If the community were to try to pass a malicious or illconceived proposal through, the multisig could call the markProposalAsInvalid() function on the SafeSnap plugin’s Reality module to veto the proposal.

Another important feature is the ScopeGuard which could be used to limit the scope of what the multisig signers and/or the SafeSnap plugin is able to do. For example, a next step might be to limit the multisig signers to only being allowed to call the markProposalAsInvalid() function on the SafeSnap plugin’s Reality module.

This solution has a couple of nice properties:

  1. it leverages the tools you already use, so there is minimal coordination needed to get it up and running.
  2. it can be implemented in small incremental steps.
  3. it doesn’t price anyone out of participating in governance.
  4. it’s flexible, composable, and portable, so you’re not making any decisions that lock you into a specific governance path in future.
  5. it gives token holders autonomy over the Gnosis Safe, while also having guard rails that can be removed as you build confidence in the system and community.

For the avoidance of doubt, the progressive decentralization pathway that I’m suggesting here is:

  1. Gnosis Safe + multisig
  2. Gnosis Safe + multisig + Snapshot (Sushi is here)
  3. Gnosis Safe + multisig + Snapshot + SafeSnap plugin
  4. Gnosis Safe + multisig + Snapshot + SafeSnap plugin + Scope Guard
  5. Gnosis Safe + Snapshot + SafeSnap plugin
  • Is the Treasury multi-sig still desirable, functioning similar to a Board?
    I’m open to hearing alternative designs, been looking at ElasticDAO’s governance system most recently. However I don’t have a problem currently with adapting the multi-sig to be the foundation for an electoral governance system for the future. I don’t see a “board” sacrificing decentralization as long as there’s healthy turnover and transparency.

  • If so, should key signers be expected to run for reelection?
    I don’t see why not! Dual benefits of allowing board members who wish to change their role to step down, while allowing community members who wish to be more active an opportunity to step up.

just some turtle thoughts


It seems worthwhile to consider converting this into a proposal if someone in the community wants to help.


Let’s make it happen! Happy to help get things set up.

1 Like