GIP-143: Should the GnosisDAO Terminate karpatkey Treasury Management Services?

[claberus]

@mrtdlgc

Now that you showed up. One genuine question. At what point the payroll fell under karpatkey itself and removed from the expenditures of Gnosis LTD?

The payroll sat on kpk from day one, funded with my own $1.5M seed. We ran at a loss until GIP-20 went live.

@filoozom

It would also be great to see concrete examples of what your business development brought to Gnosis, as this has historically been quite opaque in my opinion, not only from kpk.

Native Yield (sDAI)
We took the idle DAI sitting in the Gnosis Bridge and put it to work in the DSR. We built the agents that claim, bridge and distribute yield on Gnosis Chain. Users deposited xDAI on-chain, minted sDAI, and accessed native yield.

Chain TVL moved from 89M to 264M in under two months. sDAI has held above 100M in stablecoin market cap from Oct-2023 to Oct-2025.

Bridges
The legacy xDAI Bridge and Omnibridge were bottlenecks. In 2024 we shifted to bringing in new bridge providers. We led or supported eight integrations:
DeBridge (2M monthly volume)
Eywa / CrossCurve (0.04M)
Swaps .io (0.2M)
Relay (2.6M)
Symbiosis (0.15M)
Stargate (3M)
Bungee (1.5M)
Squid (0.5M)

DEXes
Curve was the historic venue on xDAI. Balancer and Uniswap V3 came online, reaching 1.7M and 5.3M in daily volume at peak.

Chain-wide monthly volume went from a 40M peak in 2023 to 180M in 2024, averaging above 80M, and in 2025 sits consistently above 200M with peaks around 400M.

CEXes
Ascendex was the only viable on/off-ramp early on. We delivered integrations with BuenBit and Lemon, and enabled Ripio indirectly.

Protocol Integrations
We led the following integrations:
AAVE
Spark
Balancer
Gyroscope
Aura
Uniswap V3
Gamma
Arrakis
DeBridge
Eywa / CrossCurve
Relay
Symbiosis
Stargate
Bungee
Squid
PoolTogether
Contango
Paraswap
We enabled:
PWN
Kinetex / Swaps .io
Backed
Stakewise V3
Breadchain
DELV

Agents
Operational load dropped with the Aura Harvester, Aura Relocked, wstETH Incentives, Bridge Keeper and Bridge Relayer agents.

Forex
We designed and expanded all liquidity and routing for EURe (20M monthly volume), GBPe (0.4M), BRLA (1.3M), and BRZ (0.16M).

[mrtdlgc]

@claberus

Chain TVL moved from 89M to 264M in under two months

Wasn’t this mostly due to moving the DAO assets from Ethereum to Gnosis?

[mrtdlgc]

@claberus

We designed and expanded all liquidity and routing for EURe (20M monthly volume), GBPe (0.4M), BRLA (1.3M), and BRZ (0.16M).

Also, it should be noted that especially in the case of GBPe, it took way longer than 6 months to actually establish good enough liquidity, which had a negative impact on Gnosis Pay’s growth in the UK due to people not having access to GBPe liquidity as well as not being able to still mint GBPe directly though Monerium.

[StefanGeorge]

It became apparent that the fees did not align with the delivered service, and this was mentioned as a point of discussion. KPK did, however, act proactively in this instance, reducing the fees without any concrete proposal from us. While we appreciated this move, it was also obvious that this was just one necessary step.

The following fee reductions were pushed for by Gnosis, including the latest terms.

Regarding the joint treasury, a 2% fee was discussed to foster Safe token utility and to build up a team with a clearly communicated target to review terms after 6 months. This review was not done and was not forced by Gnosis or Safe, nor was it brought up by KPK. I attribute this mostly to Gnosis and Safe having to deal with more pressing items at that time.

[Netrim]

It was really driven by sDAI, DAO assets moving from Ethereum to Gnosis Chain doesn’t matter for TVL unless they are deposited onto DeFi, thats basically the standsrd that Defillama uses and everyone has adopted. Because of economics of scale the DAO assets cannot realistically be deployed 100% onto Gnosis Chain

[NolanV]

Hey Netrim

Allow me to correct a slight inaccuracy on your part, my dear contemporary.
The increase in TVL does indeed seem to be largely due to the deployment of the DAO’s treasury.

Here are just 3 wallets, directly controlled by the DAO; it has deployed over $65m of TVL in on-chain protocols, which consequently represents >50% of the chain’s total TVL.

DeBank | Your go-to portfolio tracker for Ethereum and EVM

DeBank | Your go-to portfolio tracker for Ethereum and EVM

DeBank | Your go-to portfolio tracker for Ethereum and EVM

[aurelius]

Just had a quick look at the V3 positions on the link you shared around $8M is out of range, and it looks like kpk has never ever adjusted a V3 position?

What is this liquidity? lmao - it’s all by kpk. On wstETH/sDai

[Netrim]

I was about to edit my response because it came about like the DAO didn’t actually deposit anything.
The 5e6f wallet is the oldest GnosisDAO treasury management wallet, the two other are contemporary that came about after the ByBit experience. They didn’t exist when sDAI came about in October 2023.

For the 5e6f, it is true that SOME assets were deposited onto DAI/sDAI thru the official gnosis bridge.
This is september 2023

and this october 2023

and then november 2023

There is a marked incease on DAO assets on Gnosis Chain, yes
Is that increase the sole responsable driver for Gnosis Chain TVL growth, no
Stables were still at 11M;, then 24M and 38M, way below the 90M that sDAI had accumulated in November 2023

Edited for clarity

[NolanV]

I’d like to take advantage of this positive and friendly atmosphere to ask another question that’s been bugging me.

Why does kpk buy GNO from an illiquid pool, regularly causing volatility spikes across all markets due to the ripple effect? Isn’t it a shame to make the DAO buy at these peaks via TWAP, resulting in a very unfavorable execution for it?

Especially since ALL purchases from this pool over $5k are made by the DAO.

Aren’t you worried that the arrival of gEUR will cause bad debt problems for the CDP?

Source: CoW Explorer

[ernst]

Sorry guys, I should have spoken up earlier. Let me explain how I see the situation:

Can @ernst @StefanGeorge or @mkoeppelmann confirm that kpk proactively reduced fees, with no one ever raising this matter beforehand? It’s easy to lie about matters that aren’t publicly verifiable.

• Kpk really did reduce fees proactively on one occasion. The other two times were pushed for by us. It is also true that kpk repeatedly stepped up when Gnosis was going through a rough patch (such as the xDai team stepping away from one day to the other without prior warning, or the Builders ceasing operations). They proactively took up things that were falling by the wayside because the core team was struggling during and after these transitions.

• Kpk has been loyal to us, and I see no malice in kpk’s conduct.

• We paid Marce’s salary while he was still at Gnosis Ltd (he joined as the project manager for Open Ethereum). When he left Gnosis and started kpk, we didn’t pay for anyone’s salary. Despite the fact that (unlike for Safe and Cow) he hadn’t built kpk while still on our payroll, the DAO received 7.5% of the kpk token.

On the negatives:

• We overpaid for a significant amount of time, which was partially on us – kpk was sticking to the contract that had been negotiated (on the DAO forum) and we didn’t flag that it was no longer commensurate with service levels for too long. There were also times when I believe kpk was operating at a loss with the services it provided to us.

• Kpk 's communication has been terrible, and the execution has often fallen short.

The current proposal was renegotiated in October, but for some reason wasn’t communicated with the DAO by kpk, despite the fact we asked them to – there were some other things going on (balancer hack etc), so this might have just not been top of mind. The current proposal has clear KPIs, something that earlier proposals lacked, the glossed over most parts of the service to be provided.

In the mid- to long-term I can imagine the treasury management to be separated from strategic liquidity provisioning/ incentives on Gnosis Chain. The non-strategic treasury could be divided among several service providers for curation. The strategic parts should mostly be automated but will continue to require very close monitoring. We may have to find another service provider to take this on, or, alternatively, start doing this from Gnosis Ltd again.

In light of our history with kpk which has, on the whole, been positive for Gnosis, I would recommend staying with kpk for now while we look for a more load bearing solution in the mid-term. However, I absolutely understand the frustration felt by many of you and will do my best to find alternate solutions should this vote pass.

I am going to abstain from this vote (as are Martin and Stefan, I believe). In addition to having exposure through Gnosis, I also personally hold some kpk.

[filoozom]

@claberus

Native Yield (sDAI)
We took the idle DAI sitting in the Gnosis Bridge and put it to work in the DSR. We built the agents that claim, bridge and distribute yield on Gnosis Chain. Users deposited xDAI on-chain, minted sDAI, and accessed native yield.

So I guess you are indeed taking the credit for all of this then. From my point of view, our team at Gnosis developed the bridge smart contracts, commissioned the security audits, built the legacy sDAI frontend (admittedly with someone who now works at kpk), and manually executed yield bridging operations when your “agents” were down (for example here).

If suggesting an idea and doing the bare minimum of treasury management (in this case setting up a cron job “agent”) is the benchmark, I wonder how much I can trust the rest of this post. It’s a bit sad, because I know for a fact that you worked on a lot of these, but I’d honestly expect a bit more granularity.

[Netrim]

I have seen that image or one very similar was shared with me. Although its true there has been a buyback initative since October 23rd, anything you see before that in 2025 is not actually being done by kpk.

Having said that, there was an actual operational issue where liquidity was removed for GNO related pairs and that impacted the community ability to buy or sell GNO above 100k USD. This was corrected yesterday.
The buybacks 5k USD clip size ahouldn’t be driving the spikes as you see them so I will follow up on it.

[NolanV]

[refri]

as I said before…I am not in general unsatisfied with KPKs work, but highly unsatisfied with communication with DAO, this is not only addressed to kpk but also to ltd, cause the ties between kpk and ltd seems much closer that to DAO of any of these.

Cause now we have some statements from ltd owners I will not insist to give us some details about the discussion between ltd and kpk (without involving/informing the DAO, which I see as the main probleme) cause this is a different topic (but imo of much importance)

But I like to repeat my precise question posted here: what would be your opinion and why was proceeded as it was, and do you think this will be handled different in the future and how?

[0xAlex]

Kleros Cooperative will make its final decision based on the outcome of the Futarchy market using the Time-Weighted Average Price (TWAP) over a 24-hour window.

We believe that price-based Futarchy is an innovative and accurate mechanism for governance decisions. The prediction market has substantial liquidity, and in our opinion it is an effective way to maximize value for GNO token holders.

You can participate in and view the market here

[citrullin]

@ernst @claberus I love how you both came forward with more details. And more transparency. The only way forward here to rebuild trust you have lost. Sole reason why I am poking holes. Otherwise you will go down like any other projects before. Really enjoy the vibe here. Even though it gets heated from the community a little. There is some anger in the air. But let’s be real, that’s mostly just them looking at the market right now

How about for now and the future especially? Extreme transparency.
You don’t let 30 mil etc. disappear into nirvana, but actually just publish all the significant information.
Maybe even invite someone trusted from the DAO community itself.
Some points, from my perspective, are very valid. Especially for the risk management.
Crypto people can have a different approach to risk than what you normally would manage.
But reality is also, if the DAO is willing to take that risk. You have to go with it.
And you may have to ask yourself the uncomfortable question:
Did we foster an unsustainable community?
I don’t think that’s the point btw. One of the MOST healthy communities in crypto.
But you really have to take the leap of faith here. Overcome the afraid mindset of this ecosystem.

It’s funny really when you think about it. Trustless systems are such a myth.
There is so much trust needed and to be build in order to make it work.
But you have to take the uncomfortable jump first and just make yourself vulnerable.

And the community has to overlook the fact that humans are just humans and we do make mistakes.
Let the past be past and think about how you want to move forward.

A disclaimed from experiences, a switch could just as much get much worst than it already is.
There is no guarantee that things turn out better by any means.
With new ones, you may get a whole new set of issues and it may even take down the whole thing.
Keep that in mind when you make your decisions. Have fun out there in the trenches

[aurelius]

@claberus

after discussions with large tokenholders

[refri]

@claberus @Karpatkey

A simple ’ I agreed on what @StefanGeorge answerd in this thread’ would be enough.

Maybe you thought it was enough that he answered. And ofc it would be enough If the ommunity would be aware that a reply from ltd holds true for kpk too.

[gno-investor]

Huge thanks to everyone who voted ‘Yes.’ This is a big win with zero opposition.

We have zero tolerance for overspending at Gnosis!

Whether you are a contributor or part of the ecosystem, hit me up privately if you see overspending.

[jmo]

Hello! I see that the proposal has passed.

I’m guessing Gnosis Ltd is taking care of the transition efforts. An update regarding progress as well as the suggested next steps would be great.