Alloyed Assets on Osmosis: Unifying UX and Solving Liquidity Fragmentation

Introducing Alloyed Assets on Osmosis, the Ultimate Solution for Asset Composability


Liquidity fragmentation and user experience (UX) present significant challenges, particularly within cross-chain platforms like Osmosis. Osmosis allows users to deposit assets from an ever-growing list of bridges and source chains.

While this diversity is valuable, it also fragments liquidity and creates a UX nightmare. Osmosis’s solution is the innovative ‘Alloyed Asset’, which consolidates different versions of the same asset into a single, unified token.

Current Strategy and the Shift to Alloyed Assets

Historically, Osmosis has tried to tackle liquidity fragmentation by designating an asset from a single source as the official version within its ecosystem. For instance, ETH from mainnet Ethereum via the Axelar bridge became the canonical ETH on Osmosis, while other versions are marked with icons and affixes indicating their origin, like ETH.wh for assets deposited via Wormhole’s bridge or polygon.ETH.axl for ETH deposited from Polygon via the Axelar bridge.

While the suffix notation helps to clarify an asset’s source chain and bridging path, they don’t address the lack of fungibility between these assets for users. For example, polygon.ETH.axl (ETH deposited from Polygon via the Axelar bridge) is not fungible with ETH.axl (ETH deposited from Ethereum mainnet via the Axelar bridge), despite passing through the same Axelar bridge.

Clearly, as the variety of asset sources expands, so does the sheer volume of like-kind assets with affixes attached. So while this strategy has been an appropriate UX solution to date, it’s no longer sufficient. New developments and innovations have brought forth an even better solution–and this is where the launch of Alloyed Assets comes in–as the end-game vision for seamless composability of like-kind assets on Osmosis.


Taking Inspiration from Centralized Exchanges

Osmosis drew inspiration from Coinbase’s streamlined system for managing assets from multiple sources. On Coinbase, all deposited ETH, regardless of its origin, is uniformly labeled as ETH, greatly simplifying the deposit and withdrawal process.

People that use centralized exchanges (CEXes) rarely consider the custody solutions used under the hood to secure their assets. On Osmosis, bridges are intended to play a similar role to CEXes’ custody services, simplifying asset transfers while masking the underlying complexity, and providing a seamless experience for users.

Alloyed Assets: A Novel Solution

Osmosis is revolutionizing how like-kind assets are managed with the introduction of Alloyed Assets. This system leverages the ‘transmuter’, a CosmWasm smart contract that implements a stableswap AMM (optimized for like-kind assets, to enable 1:1 swaps with no slippage).

Alloyed assets represent LP shares of these pools, combining various asset versions into one fungible token. Depositing different asset versions into the pool adds liquidity and, in return, users receive LP shares in the form of alloyed assets, enhancing UX by consolidating liquidity.

Users can acquire alloyed assets by simply swapping a like-kind asset for the alloyed version in the swap page.

Alloyed assets are optional for users. Anyone who wants to hold the non-canonical, single-bridge version of an asset is free to do so.

The Great Bridge Fungifier

In a sense, the Alloyed Assets system acts as a bridge aggregator or ‘bridge fungifier’, ensuring assets transferred across diverse bridges become fungible within the Osmosis ecosystem.

Osmosis governance will determine which assets will be included in each Alloy (i.e., Alloyed pool).

Two-Tiered Approach to Managing Risk in Alloyed Asset Pools

Osmosis leverages a two-pronged rate limiter mechanism to: 1) maintain balance within Alloyed asset pools; and 2) ensure that if one source bridge is compromised, the remaining assets in an Alloyed pool from other bridges are not at risk:

  • Static Limiter: This static limiter mechanism is a safeguard that prevents any single version of an asset from dominating the pool’s composition beyond a predetermined threshold.
    • For example, an ETH alloy can limit ETH via the Wormhole bridge from exceeding 30% of the alloy’s composition.
  • Change Limiter: On the other hand, the change rate limiter mechanism is intended to monitor and control the rate at which an asset’s share of a pool can increase over a given timeframe. This provides a window for oversight and potential intervention, ensuring the pool’s stability and integrity.
    • For example, an ETH alloy can limit ETH via the Wormhole bridge from increasing its composition by more than 10% in a 24 hour period.

While Alloyed Asset pools increase a user’s exposure to more bridge-related risks, the risks associated with each potential bridge hack or failure are lessened via the rate limiting mechanisms described above.

For additional safeguards, the system will employ a SubDAO that can quickly respond to any incidents or anomalies. If an asset within an Alloyed pool is deemed corrupted, the SubDAO can mark it, triggering a mechanism where the corrupted asset’s amount and weight in the pool can only decrease. This prevents it from dominating the pool by requiring users to redeem corrupted assets in proportion to non-corrupted ones, to maintain balance.

The system also allows for the possibility of an altruistic actor to redeem all corrupted assets in a singular transaction, bypassing the limiters to cleanse the pool of the corrupted asset. Following the clearance of corrupted assets, the pool resumes normal operations, possibly with adjusted limiter settings to reflect the new asset composition. This streamlined approach ensures the integrity of the pool and protects against value depreciation.

USDT and Beyond

USDT is anticipated to be the first alloyed asset proposed to use the Alloyed Asset system—the first of many. You can get involved in the discussion here: Alloyed Asset: USDT

With alloyed assets, Osmosis aims to redefine liquidity management and UX in DeFi. This marks the beginning of an ambitious journey to support the deposit and withdrawal of a wide range of assets from diverse sources.

Osmosis is committed to continuously improving and expanding its Alloyed Asset framework. By directly addressing liquidity fragmentation, Osmosis not only enhances its platform, but also contributes to the overarching goal of seamless blockchain interoperability.

FAQs

Do I have to use an alloyed asset?

No, alloyed assets are optional. If a user wants to hold a specific version of an asset, that will be possible. However, the UX of osmosis.zone and likely other frontends will be optimized for the canonical assets, and pending governance approval, liquidity will be directed towards alloyed assets via incentives and market maker initiatives (as in prop 712).

What happens if an underlying asset runs out in an alloy?

If an underlying asset runs out in an alloy, that withdrawal path is no longer usable until refilled. In the future, Osmosis governance can choose to incentivize such rebalancing to maintain minimum liquidity of different underlying assets.

Will LSDs be included in the alloys?

No, because LSDs are not designed to maintain price parity with their underlying assets, they will not be able to be included in alloyed pools. However, there could be alloys for specific LSDs – for example a wstETH alloy enabling bridging of wstETH from numerous Ethereum L2s.

Will Rebasing tokens be included in the alloys?

No, Osmosis pools and IBC in general are not currently designed to support rebasing tokens, and they will also not be usable in alloys.

Will different stablecoins like USDC and USDT be included in the same alloy?

No, because different stablecoins have different backing and risk structures, and have historically deviated off of a perfect peg.

Can Alloyed Assets be transferred?

Yes, they can be moved within the Osmosis ecosystem and beyond in the same manner as other IBC-enabled assets. This means users can transfer their Alloyed Assets between wallets and chains, and utilize them in any protocols that list or support them. For now, the full functionality of withdrawing the individual underlying assets only exists on the Osmosis DEX.

How do users and protocols interact with Alloyed Assets compared to standard assets?

Users and protocols interact with Alloyed Assets similarly to standard assets in terms of basic functionalities such as transferring, trading, and utilizing them in DeFi protocols. However, the key difference lies in the abstraction of bridge details and asset composability facilitated by Alloyed Assets.
Users transacting through the app.osmosis.zone frontend will have the management of bridge connections and liquidity abstracted away through an intuitive withdrawal and deposit process.

Protocols may need to integrate specific functionalities to handle Alloyed Assets, such as utilizing Skip’s API to handle the routing between variant assets that a user may wish to deposit and the protocol’s choice of using the Alloyed Asset or specific constituents.

Chains may choose to use the Alloyed Asset as a new standard, using Osmosis DEX as a hub for deposits and withdrawals in the same way that bridging chains currently are used as such. They may also implement a mechanism similar to Alloyed Assets and unwrap/rewrap the constituent assets using IBC middleware during the transfer process.

9 Likes

I generally feel pretty bearish Alloyed assets.

On a CEX, you have a single counterparty that assumes all the fungibility and bridge hack risk.

While the mitigations do provide some buffer against catastrophic failure, I think LPing in an alloyed asset pool make me pretty uncomfortable.

I also worry a lot that adding an additional click/signature to common user flows like swap then bridge and introducing complexity where liquidity routers like skip/tfm will have to alloy/unalloy assets in their workflows.

Just some feedback.

The liquidity fragmentation is unfortunate but I’m not comfortable that alloyed assets are an amazing fix.

2 Likes

Well I certainly hope that alloyed assets will be in addition to as opposed to instead of.

I would hope that users can still access the underlying assets.

Yes, underlying assets will still be usable in pure form.

Including for in LP pools. Just as now, you can continue to LP in the OSMO/USDC.axl pool, post canonical ETH becoming the alloy, you will still be able to LP in ETH.axl pools.

Alloyed asset pools are implemented as cw_pools, meaning that to alloy an asset just requires doing a swap (similar to the existing transmuter pools). For example, going from ETH.axl → ETH is just a swap. As most APIs like Skip API and Squid already support triggering a swap immediately upon bridge deposit without need for an additional tx, from a user perspective this will be abstracted away.

The osmosis.zone frontend dev team is currently doing the same for the Osmosis site, that you will be able to deposit ETH via one of the bridges, and have it automatically alloy the asset using ibc_hooks .

4 Likes

OH YEAH!

Having users use the alloyed one automatically is so much better for the less-/non cryptonatives.
And adding a checkbox if people want to use a native asset for the pro-users.

But making UX better to abstract away from the prefixxes; everyday.

3 Likes

I doubt this idea hasn’t been contemplated by the team already, but for further risk isolation/reduction, what about:

I deposit polygon.ETH.axl into an alloyed pool which is mixed with ETH.wh.

polygon.ETH.axl gets corrupted, the subDAO triggers corrupted status for that alloyed asset pool. The system remembers how each LP provider entered the pool, and when a pool is in corrupted state, the user can only exit the pool via the same asset they entered (or in the same proportion as they entered, if they entered via more than one asset), in my case polygon.ETH.axl.

There might be advantages or disadvantages to applying the same restriction (one exit lane) to non corrupted assets, in the case of a pool with 3 or more assets. In a 2 asset pool, the restriction is logically enforced because there are only 2 options, and no one who entered with ETH.wh will want to exit with polygon.ETH.axl. But this is a lesser point at this point in the conversation.

If this is feasible, it would really help reduce risk. because as is, the risk of damage for any given alloyed pool is much higher than the individual corruption risk of each asset entered in the alloyed pool.

Asset A risk of corruption: 5%
Asset B risk of corruption: 3%
Asset C risk of corruption: 15%

Alloyed pool risk of corruption: 21.63%

PS: If the math above is incorrect this idea becomes less relevant.

2 Likes

Doesn’t change too much but I got 21.52% for the risk of corruption when calculating the probabilities.

1 Like

…surprised there was no reply on this objectively (higher) risk option, and what could be done to reduce it.

While there is a higher chance of the Alloy suffering any corruption event, the impact of any corruption is hugely reduced thanks to the limiters. This is the key to Alloys—we aren’t putting all of our eggs in one basket anymore so that chance occurring is not as critical, but the chances of a corruption event occurring will of course rise with a greater selection of bridges within the alloy.

When setting these limiters we can use this risk evaluation to set the tolerance for these to attempt to restrict the impact, for example setting Asset A at 60%, B at 100% and C at 20%:

Asset A impact of corruption: 5% * 60% = 3%
Asset B impact of corruption: 3% * 100% = 3%
Asset C impact of corruption: 15% * 20% = 3%

If a bridge is seen as risky, then it should be less likely to be trusted for transfers, and so the static limiter is even less likely to be hit and the impact controlled. To the point where a holder of an Alloy is incentivized to deliberately remove the risky asset from the Alloy by arbitrage to improve their own risk exposure whilst avoiding the total loss that may result from holding a single bridged asset.

If corruption occurs, withdrawal is limited to anything but the corrupt asset—the opposite of what you’re saying is that users who deposit the corrupted asset would take on all of the risk of that deposit mechanism and be unable to withdraw from the pool. This could cause a lot of confusion during a corruption even for users who hadn’t realised they only deposited via the corrupted bridge and are now trapped within the Alloy.

Generally, the resolution of corruption events will come down to compensation of the Alloy by either the bridging provider that failed or Osmosis governance - and the limiters restrict the cost of this for both parties.

1 Like

Voted ! One click with Osmosis allASSETs !!
With six finger’s !!