ATOM wars: Hydro pilot rounds

Edit 18-Oct: After consultations with a number of Cosmos contributors & security experts, the launch of Hydro, initially planned for 16-Oct, is temporarily paused. This follows the news that North Korea has infiltrated a number of crypto projects and that NK developers have contributed to the LSM module. Informal had no knowledge of this prior to the Coindesk article and Hydro currently uses LSM shares. The team is now working on a replacement solution and facilitating audits. We hope to provide an update on a new launch timeline next week

Hydro’s initial rounds will be “pilot rounds.” Users will be able to lock their staked ATOMs to get voting power and projects will be able to bid to receive a share of a single liquidity bucket seeded with 200,000 ATOMs.

The goal of the pilot rounds is to identify and fix bugs, gather user feedback, refine the auction process and establish a baseline. It is a learning opportunity for projects, users, the Hydro committee and the Informal team.

In this post, we explain the pilot rounds, the Hydro committee guidelines, the exact launch sequence, the status of the Prop955 claw-backs, why Hydro v1 is launching on Neutron (instead of directly on the Hub) and Informal’s ideas on the LSM integration & Hydro v2.

Committee guidelines

Hydro is a first-of-its-kind product. As such, and particularly at launch, the system requires active monitoring & decision-making to ensure that liquidity deployments are profitable, and align with the strategic goals of the Hub while protecting the interests of all participants. It will take a lot of incremental improvements to get to a fully automated, risk-minimizing & profit-maximizing system. At launch, Hydro relies on a committee for the following functions:

  • Whitelist management: It maintains a whitelist of projects based on objective risk mitigation and/or hard criteria, primarily based on on-chain market data. These are communicated to the projects under review, ensuring a transparent process
  • Export performance: It monitors the performance of all liquidity exports, ensuring that all participating projects utilize the funds in alignment with their stated purposes and achieve the anticipated returns.
  • Auction performance: It monitors bidding strategies and identifies potential manipulations or cartelization efforts. It may use tactics such as analysis of bidding patterns and unusual fluctuations in bid amounts; and audits of winning bids.
  • Emergency warnings: It identifies any manipulation or malicious behavior and is responsible for initiating a proposal to the Cosmos Hub to potentially inflict a monetary sanction to malicious voters

In order to make the decision-making process on the above items as transparent as possible, the committee is publishing its initial guidelines in the below document. These guidelines are meant to evolve over time. The committee is adopting a very risk-averse strategy at launch, as the initial goal is simply to improve on the current method of PoL deployments and experiment with a new approach. The consensus among committee members is that Hydro should restrain from venturing too far to the right on the risk-profit curve for the first couple of months.

Hydro committee guidelines

The committee has also been working toward establishing a Conflict of Interest policy, a first step towards codifying key processes such as adding or removing members, ensuring smooth transitions, maintaining transparency and fairness etc. The document below is a starting point and as these procedures get implemented, numerous areas for improvement will inevitably be identified. Community input will be valuable as we collectively refine these guidelines.

Hydro committee CoI policy

During Pilot Rounds, the bid approval process will be done via a public Github:

  • Projects will submit their bid details via GitHub
  • The Hydro Committee reviews the bid details and leaves a review
  • Approved bids get submitted to the Hydro contract from a whitelisted address

Once the bid is on-chain, projects can start adding tributes to their bid

Any project interested in receiving more information may email hydro@informal.systems. A detailed project onboarding guide will be provided on the Hydro website at launch.

Pilot rounds rules

Committee management

Hydro’s smart contract has a Top-N feature controlled by the Hydro committee. The goal of this feature is to avoid exporting small winnings, which in the absence of the Valence integration can be very time-consuming (each deployment requires the Hydro committee to coordinate and sign a series of messages).

During the first pilot round, N will be set to include projects that receive 10% or more of the votes. Since the bucket starts with 200K ATOMs and ATOM is around $5, this will roughly correspond to a minimum of 100,000 USD for exports.

The liquidity available in the bucket will be distributed among these top N projects and any projects that do not make it into the top N will have their tributes refunded.

Capped buckets

Per Prop 955, the total amount of liquidity re-routed to Hydro is expected to be close to 2.5M ATOM. This is without counting the protocol-owned rewards which were generated during the POL deployments.

The transfer of custody from various multisig to the Hydro committee should happen before the end of the pilot rounds (see section below for a full status update).

During the pilot rounds, Hydro will use a smaller bucket seeded with 200K native ATOMs re-routed from Persistence. The Hydro committee is not planning to exit from the existing positions until the liquidity is re-deployed by Hydro voters.

Voting power

During Hydro pilot rounds, users can only lock their staked ATOM for the duration of the round (1 month). This is meant to reduce risk around longer locks. Once regular rounds start, users will be able to pick longer locks, which will give them additional voting power.

Tranches

The absence of an ICS tranche during pilot rounds will simplify the flow for users. It will also reduce the workload of the Hydro committee, who is currently operating on a volunteer basis.

Clawbacks status

Informal is coordinating with various ad-hoc multi-sigs to transfer custody of funds to the Hydro committee, so it can be re-allocated by Hydro voters.

Persistence: The Prop 853 multisig is transferring 600,000 ATOMs. The first step was completed on 4-Oct and the second step (stkATOM un-staking) is in progress.

Osmosis: The Prop 858 multisig is transferring custody of 900,000 ATOM. Proposal A22 is in progress, awaiting signing from members.

Stride: The Prop 800 multisig had custody of 450,000 ATOM. The funds were sent to Valence per Prop 950. A separate proposal is required to transfer custody to the Hydro committee.

Mars: The Prop 946 multisig will be transferring custody of 200,000 ATOM from the AADAO to Hydro committee at the end of the first pilot round.

Agoric: The Prop 912 multisig is yet to transfer custody of 188,768 ATOM.

Launch sequence

In the table below, we summarize the upcoming Hydro launch steps:

# Stage Description
1 Projects start submitting tributes Once Hydro launches, projects will begin submitting their bids for liquidity along with the tributes they’re offering to incentivize ATOM holders.
2 Users lock their LSM shares and vote With the first round of projects live, users can start locking up their ATOM to gain voting power and cast their votes.
3 Pilot round 1 tributes are distributed to voters At the end of the first pilot round, tributes are distributed proportional to the amount of LSM shares that voted for each project.
4 Liquidity is deployed by the committee Liquidity is deployed proportional to the amount of LSM shares that voted for each project. This will likely take a couple of days
5 Projects start submitting pilot round 2 bids New projects submitting bids & adding tributes. Round 1 voters may refresh or extend their locks. New users can lock new ATOMs
6 Pilot round 2 tributes are distributed to voters At the end of round 2, tributes are distributed again, and the cycle continues with more project bids and more voting opportunities

Hydro & Neutron

The Informal team added CosmWasm to Gaia in early July with the v18 upgrade so launching on the Hub was absolutely an option that was considered seriously.

First, it’s important to understand that the Cosmos Hub is currently using a permissioned version of CosmWasm, which means every contract deployment must first go through governance. This would have delayed the launch by another month (two weeks on forum, two weeks on-chain) at a time where many projects are impatiently waiting for Hydro to launch.

More importantly, we hope to see a permissionless ecosystem of many DeFi projects built on-top of Hydro, for instance:

  • Vote aggregator that abstract the complexities of voting away from users, give them the benefits of long-term locking, provide a tradable liquid derivative, compete for maximal returns etc.
  • Perpetual future contracts to avoid exposing the Cosmos Hub deployments to impermanent loss. Hydro’s bidders would be able to maintain a delta-neutral strategy to hedge, with a real-time adjustment mechanism via smart contracts.
  • A covenant protocol (Valence) that ensures that the Hub retains direct ownership of the liquidity, eliminating the need for a committee to manually deploy funds and increasing the overall security of the system.

The Informal team is already in touch with third-party engineering teams on these items. It’s also likely that creative DeFi teams will come up with new ideas that we haven’t considered yet. Because Neutron has permissionless smart contracts, it allows for easy & fast building without the added IBC complexity or governance approval.

The decision to deploy Hydro v1 on Neutron also led us to use DAO_DAO on Neutron. This will make it easier for the Hydro committee to interact with the Hydro contract without a dependence on relayers & without delays. For instance, the DAO would still be able to pause the contract even if the Hub to Neutron IBC channel is down. Security is a important consideration. We prefer to minimize the moving parts by having everything on one chain. Neutron is a CosmWasm gold subscriber and it has the expertise and operational capacity to respond to security issues.

Neutron is also the permissionless CosmWasm environment of the Hub, sharing 25% of its revenues. Neutron accepts ATOM as a gas token and the Hydro frontend will default to using ATOM for gas. With all that said, we still made a deliberate effort to design Hydro in a way that makes it relatively easy to port it to another chain, and a new trade-off analysis may lead to Hydro v2 or v3 to be deployed on the Hub instead.

Hydro & LSM

ATOM stakers will tokenize their stake via the Liquid Staking Module (LSM) to get voting power in Hydro. They will then use this voting power to allocate Protocol-Owned-Liquidity.

The LSM is a form of regulation on liquid staking providers : it limits the total amount of ATOM that can be liquid staked and requires that validators self-bond ATOM to be eligible for delegations from liquid staking providers. The LSM also allows for staked ATOM to be liquid staked without having to wait an un-bonding period.

Informal Systems is supportive of the LSM and voted yes to Prop 790. However, the current implementation of the LSM has several issues:

Maintenance: The LSM is not part of the mainline Cosmos SDK. The Hub uses a custom SDK branch for LSM support. The LSM was written for the SDK v0.45 and the Informal team had to spend time upgrading it v0.47 and v0.50. This will continue to be a requirement until the LSM logic is made part of the Cosmos SDK. The LSM adds complexity on top of the complexity of the staking & distribution, which increases the likelihood of production bugs.

Security: A Coindesk article from 2-Oct revealed that the original version of the LSM was partly developed by North Korean IT workers. Although the LSM was later re-written by Iqlusion & Stride (and the new version is currently being audited by Oak Security), we can’t completely discard the risk of a planted vulnerability or a backdoor.

Voting: It is currently impossible to use tokenized staked ATOM to override your validator vote in Cosmos Hub governance. It is possible to develop a new Hydro feature that would let Hydro LSM lockers submit their votes via Hydro governance and then use a weighted vote over Interchain Accounts to vote on Hub governance, but this is not an ideal long term solution.

Integrations: Wallets like Keplr & Leap do not yet display tokenized staked ATOM nicely. Although locked LSM shares do accrue staking rewards, this is not made obvious in the wallet interfaces and users may wonder where the staked ATOMs went. In addition, it is still unclear how airdrops scripts will treat LSM shareholders.

The Informal Hub team has found a solution to the LSM shortcomings. In short, the user would give the Hydro contract permission to withdraw their staked ATOM on their behalf. This is similar to the way Eigenlayer is designed, but it would require significant updates to the Cosmos Hub’s SDK staking module.

Hydro v2

“If you’re not embarrassed by the first version of your product, you’ve launched too late” - Reid Hoffman

Hydro v1 is a minimal viable product:

  • It relies on a committee for whitelisting & deployments
  • The incentive alignment between voters & the community pool is weak
  • The round duration is fixed

The experience gained from deploying v1 will shape the development of Hydro v2, which we expect to be a much more robust iteration of the platform. Hydro’s endgame is to be the first community-governed Protocol-Owned-Liquidity allocator: a fully permissionless, automated, risk-minimizing & profit-maximizing system for efficient treasury management on a public blockchain. Hydro should ultimately be able to operate like Aave, Compound or Curve with risk parameter approvals via governance votes, allocations via an auction system and automatic deployments via Valence.

This will likely take a number of iterations to get right. For the second iteration, here’s a list of the features that we currently have planned:

Feature Description
Custom bid duration Projects are able to set their own export duration and voters are therefore able to lock-in a fixed APR over several rounds without having to vote every time
Slashing & rev-sharing A portion of the users’ locked capital gets slashed when a project underperforms. Voters get a portion of the PoL yield distribution
Hydro governance Hydro users can vote on governance proposals to define all of the protocol parameters and elect the Hydro committee
New staking interface Hydro users can stake, delegate, re-delegate, claim & compound rewards directly from the Hydro interface
Funding rate End-of-round value transfer to guarantee parity of return for voters of “low-yield high-tribute” project and voters of “high-yield” low-tribute” projects

Hydro v2 will also bring a set of improvements on existing features, such as the ability to relock all LSM shares at once, the ability to do fractional relocking, a random auction end time, weighted voting, a performance dashboard for all PoL exports and likely more.

Conclusion

Over the next couple of months, the pilot rounds should provide the Hub’s community with a good sense of Hydro’s long term potential. It will be able to decide whether or not it wishes to integrate Hydro further and make Protocol-Owned-Liquidity a key part of its strategic goals.

Further Hub integration would open up a number of options:

  • A transition to an inflation model where rewards are increased for longer locks (users who lock up their tokens for extended durations would not only gain more voting power in Hydro, but also receive extra inflationary rewards)
  • A recognition that not all ATOM holders want to participate in governance and an extension of the decision scope of active Hydro voters beyond liquidity deployments (into funding proposals, subDAOs members elections etc.)
  • A diversification of the Hub community pool into other tokens (USDT, USDC, NRTN, STRD etc.) and the deployment of the new liquidity buckets via Hydro.

On the other hand, it is also possible for Hydro to continue as a more neutral Interchain project and pursue a strategy of integration with other community pools, foundations and private capital.

Either way, as a first-of-its-kind product made possible by Cosmos’s pioneering work in interoperability and sovereignty, Hydro is working towards a future where efficient resource allocation and decentralized governance can go hand-in-hand.

8 Likes

Babe wake up

New post from @Thyborg

3 Likes

If a user has LSM module disabled for security purposes @Thyborg, does that mean they wont be able to participate in HYDRO auction bids?

I have personally disabled the LSM module through @cosmosrescue, but made me think now that I might not be able to participate in HYDRO anymore…

3 Likes

Correct, the current version of Hydro uses the LSM, so you’d have to re-enable that to participate.

3 Likes

Damn! It takes 21 days, so i guess im out of first weeks of Hydro.

Missed out i guess…

FYI, this shortcoming of LSTs is currently being solved for - working with Numia, we are building a self-service snapshot for projects, helping them easily include ATOM LST holders in their airdrops. Currently, potential leads for this service include Nillion, Landslide, and roll-ups on Celestia.

Does this mean Hydro is doing away with LSM entirely? We as ATOM holders directly lock up with the Hydro contract?

3 Likes

Note that offering only ATOM-LST ATOM liquidity (Public Goods), and at that only 33% of a pool, which means it has to be crowdfunded/sourced anyways, almost entirely nullifies the use of this liquidity for a DEX, and also forces a concentration of liquidity on more established DEXes.

There’s so much that Astrovault would like to do with Hydro. Like use our Time Flux Pools to offer high APR and be the best place to trade ATOM, and collateralizing other assets to back potential losses.

As a potential user for Hydro, this sucks. I really fear that we’re wasting a ton of time and energy to pay money to get marginal amounts of nearly useless liquidity, but on top of that that any request for USEFUL liquidity to the community pool will get shut down because its expected that Hydro can handle ‘liquidity’. There seems to be a gap between perceived utility of Hydro and actual utility of Hydro, and it’s a problem.

2 Likes

I would have preferred to keep my ATOM natively stake and just do a proof of ownership and lock mechanism without having to rely on LSM

1 Like

Snapshots: Nice, very useful!

LSM: We’re planning to keep the integration at least for the duration of the pilot rounds. After that we’ll need a new prop to confirm that the Hub wants to integrate Hydro further into the chain, and part of that prop may include doing away with the LSM (our solution above still needs more careful consideration)

3 Likes

Hydro is taking a conservative approach for the pilot rounds. There’s also a lot more we’d like to do and I hope it 'll be moving to the right of the risk curve consistently over the next few months.

Pilot rounds are just an opportunity for the Hydro committee and projects (hopefully Astrovault) to get a sense of the process & bidding dynamics (and for Informal to squash any remaining bugs)

2 Likes

Very much agree on that, - if its stATOM, then its not ATOM! Done. LSM or whatnot, the best practice would be POO - Proof of ownership or whatever and we lock up our atoms natively instead of LSM

I absolutely appreciate this mindset, and think that’s totally fair.

Just note that if shipping takes some time, the burdens of growth could cripple deal flow in the intermediate time. Want to make sure all perspectives are understood, as we all want this to work, but working under the assumption that it will may create a significant dependency for network dealflow.

4 Likes

In view of these findings from AiB, and presuming the findings are accurate, Hydro should be paused.

  • “Majority of LSM code written by North Korean agents”
  • “Alarmingly, the same North Korean developers who had contributed the majority of the original code were tasked with addressing the findings raised by the audit.”
  • 'LSM is not a standalone module but rather a series of modifications and extensions built on top of the existing Cosmos staking modules. The term “module” is more of a conceptual shorthand, used to describe the liquid staking features added to the core Cosmos SDK staking functionality. Consequently, any vulnerability in Iqlusion’s LSM that impacts these core modules could potentially put all staked ATOM at risk, as liquid staking interacts directly with staked assets."

The ICF also bears responsibility here for lack of oversight wrt to development it funds.
I also don’t understand why Informal didn’t push for an audit prior to integration with Gaia in Sept 2023.

3 Likes