mUSD as a USDe Backing Asset

This report evaluates MetaMask USD (mUSD) for inclusion in Ethena’s USDe backing, and specifically the proposed structure in which Ethena would hold mUSD and supply it into the Aave v3.7 instance on Monad to earn yield. It covers the issuer and legal structure, reserve composition and custody, smart-contract and technical risk, market data and peg behavior, the MetaMask Money Account that anchors mUSD demand on Monad, the mechanics of the Aave Monad deployment, and mUSD’s fit within Ethena’s backing framework.


Source: LlamaRisk, July 15, 2026

1. Summary and recommendation

mUSD is a fiat-backed stablecoin issued by Bridge Building Inc. (a Stripe company) on M0 Protocol infrastructure, backed one-to-one by cash and short-term U.S. Treasury bills. Its reserve profile is conservative and matches assets Ethena’s Risk Committee has already approved or assessed as backing (USDtb, USDG, and the existing USDC allocation).

The reservations concern scale, liquidity, and residual centralization rather than the reserve. mUSD is small (about $32M total supply, of which about $9.6M is on Monad) with no centralized exchange market, and its onchain exit is a single Uniswap v4 pool per chain absorbing roughly $2M near par. Issuer redemption is one-to-one but permissioned, KYC-gated at the fiat ramp, and multi-day, so it does not provide instant exit capacity. Two administrative roles on the token, account freeze and forced transfer, are assigned to single externally owned accounts (EOAs), which we flag as the main unresolved access control risk.

LlamaRisk supports onboarding mUSD as a USDe backing asset under a phased, capped allocation, and supports deploying that mUSD into Aave Monad. Initial allocation should be sized to what mUSD’s redemption throughput (as underlying $M is backed by Superstate USTB) can support rather than to Aave’s headline supply cap, and increased as supply, liquidity, redemption performance at par, adoption are demonstrated at size.

2. Issuer and legal structure

2.1. Issuer

mUSD is issued by Bridge Building Inc., the stablecoin-infrastructure subsidiary of Stripe, registered with FinCEN as a Money Services Business (NMLS #2450917). In February 2026, Bridge received initial approval for a U.S. national bank trust charter, an escalation beyond MSB registration that, if finalized, would place issuance and reserve custody under federal bank-trust supervision. Bridge issues mUSD through Stripe’s Open Issuance platform and positions the token as GENIUS Act aligned, though it does not yet hold a completed federal permitted-payment-stablecoin license. The token was announced in August 2025 as a joint effort between Stripe’s Bridge and M0.

2.2. Token architecture

mUSD is built on M0 Protocol. M0 mints a shared base token, $M, against short-term U.S. Treasuries, and approved issuers wrap $M one-to-one into branded tokens through extension contracts. mUSD is a MYieldToOne extension, meaning all reserve yield accrues to a single recipient rather than to holders, with added pausability, account freezing, and forced-transfer capability documented in the Consensys Diligence audit. On Ethereum, the mUSD contract holds roughly 15.04M of $M as collateral against its Ethereum supply, approximately one-to-one; $M total supply is about 300.6M, so mUSD is a small extension within a larger issuance network. The token deploys at the same address, 0xacA92E438df0B2401fF60dA7E4337B687a2435DA, on Ethereum, Linea, and Monad.

2.3. Holder rights and redemption

mUSD is permissionless to hold, transfer, and swap, but primary issuance and redemption are permissioned. Only Bridge, as the whitelisted issuer, mints and redeems against posted Treasury collateral. End users redeem to fiat through Bridge or authorized affiliates and compliant on-ramps rather than directly from the contract, per Bridge’s issuance FAQ and MetaMask’s mUSD support page. Redemption is at par, subject to KYC at the ramp, and settles on rail-dependent timing (ACH is typically one to two banking days per Bridge’s settlement cutoffs). There is no published mUSD-specific redemption SLA, so the roughly two-business-day figure should be read as an approximation of the standard bank-payout rail rather than a contractual commitment. mUSD currently operates under Bridge’s generic stablecoin terms.

2.4. Yield treatment

mUSD pays no interest to holders. Reserve yield accrues to the issuer side via the MYieldToOne recipient, which on-chain is an unlabeled EOA; no public source formally confirms it as the MetaMask or Consensys treasury, so that attribution is an inference from the contract design. The separation is deliberate: under the GENIUS Act a payment-stablecoin issuer cannot pay interest to holders, so decoupling the token from yield is the compliant design. It is also consistent with every fiat-backed stablecoin Ethena already uses, since USDC, USDtb, and the recently assessed USDG all direct reserve yield to the issuer rather than the token holder.

3. Reserve composition and custody

3.1. Reserve assets

mUSD’s economic backing is the collateral held against $M: cash and short-term U.S. Treasury bills in regulated custody, layered as mUSD to $M to Treasury collateral (illustrated in the stack diagram above). M0 gates $M minting to verified collateral through Chronicle Labs, which operates an independent validator that verifies off-chain reserves on-chain before mint capacity is extended. This on-chain collateral verification is the principal transparency mechanism for the backing.


Source: Chronicle Labs, July 15, 2026

3.2. Custody

Reserve collateral is held in segregated custody on the Bridge side of the structure. Because Bridge is a FinCEN-registered MSB now pursuing a national bank trust charter, custody sits with a U.S.-regulated issuer, and a finalized trust charter would strengthen the custodial framework further. Bridge does not publish a per-custodian allocation for mUSD reserves, which is a disclosure gap relative to the custodian-level detail Paxos provides for USDG.

3.3. Attestations

The main reserve-transparency gap is the absence of a named third-party attestor and a published attestation cadence for mUSD specifically. MetaMask’s own reserves explainer concedes that independent third-party verification is still developing. For an asset intended to sit inside USDe’s backing, LlamaRisk would expect the monthly attestation discipline from a recognized firm that USDG reached through its KPMG engagement and that USDtb achieves via BlackRock BUIDL and Securitize reporting. Until that exists, reliance falls on Chronicle’s onchain validation and the integrity of the M0 collateral framework.

4. Smart contract and technical risk

4.1. Contract architecture and access control

mUSD’s administrative surface is defined by its MYieldToOne extension and the M0 role model, as documented in the Consensys Diligence audit and readable on the token contract. The token exposes a freeze role that can block any account and a forced-transfer role that can then move a frozen account’s balance. Both support Bridge’s OpenSanctions compliance program, under which 146 addresses were frozen across Ethereum and Linea at last count. LlamaRisk’s concern is the key custody of these two roles: each is held by a single EOA, with no multisig and no timelock, so a freeze and seizure can execute instantly and unilaterally. For an Ethena backing position the sanctions-freeze angle matters less than for retail holders, since Ethena’s own mUSD would not be a target, but the concentrated key custody remains a live centralization vector.

4.2. Timelock and upgradeability

Role-admin and proxy-upgrade authority sit behind a 72-hour timelock, with the timelock proposer, executor, and canceller held by a 3-of-5 multisig and no open public executor. This governs the highest-impact administrative changes and is the appropriate control for upgrades and admin transfers. LlamaRisk views the timelocked upgrade path as adequate.

4.3. Cross-chain bridging

mUSD uses the M0 Portal in a hub-and-spoke design: Ethereum is the hub (lock-and-release, the sole minting and governance chain) and the other chains are spokes that mint and burn against hub messages. For the Monad and Linea spokes, message delivery runs over Hyperlane (the M0 Portal Lite path). Wormhole Native Token Transfers is the M0 Portal Standard path used for Solana and other non-EVM targets, and Monad’s own native asset bridge separately uses Wormhole, but mUSD’s Monad spoke itself routes over Hyperlane. There is no rate limiter or cap on the mint path, so bridge mint capacity is effectively unbounded. On Monad, the bridge operator and pause roles sit on a single EOA that can re-point the trusted remote, upgrades run through a 3-of-5 multisig with no timelock.

4.4. Audits

The mUSD token was audited by ChainSecurity (two informational findings), Consensys Diligence (three low, five informational), and Guardian Audits (one informational), all in August 2025, with reports collected in the M0 mUSD repository. The M Extensions, covering MYieldToOne and the SwapFacility, were audited by ChainSecurity. The M0 core protocol carries a broad 2024 audit set (Certora, ChainSecurity, OpenZeppelin, Quantstamp, ThreeSigma, and others), and the M0 Portal bridge was audited by Adevar, Halborn, Sherlock, and ChainSecurity with no open critical or high findings.

4.5. Bug bounty

mUSD is in scope on MetaMask’s HackerOne program (smart contracts and APIs) since December 2025. The rewards are ETH-denominated, so the “$50K maximum” figure cited in earlier coverage should be read as indicative rather than a fixed cap.

5. Market data and peg stability

5.1. Supply and distribution

Total mUSD supply is about $32.2M, split Ethereum $15.0M (46.7%), Monad $9.56M (29.7%), and Linea $7.57M (23.5%). Over the trailing month total supply softened slightly even as Monad ramped, so the network is concentrating weight on Monad rather than growing rapidly in aggregate. The launch-era balance had been concentrated on Linea; that has largely unwound while Ethereum roughly tripled and Monad went from zero to nearly 30% of all mUSD in a matter of weeks.


Source: LlamaRisk, July 15, 2026

Holder count is broad, at about 244,000 addresses on Ethereum per Etherscan and roughly 273,000 across chains. Per-holder concentration is moderate: the largest Ethereum holder is about 7.8% of that chain’s supply (roughly 3.6% of the global total), the top five are about 27% of Ethereum supply, and several of the largest balances are DEX or protocol contracts rather than beneficial holders.

5.2. Growth on Monad

Monad supply went from about $0.95M in mid-June to about $9.56M by July 15, a roughly ten-fold increase in four weeks, with the inflection at the June 30 Money Account launch and the July 2 Aave on Monad deployment.


Source: LlamaRisk, July 15, 2026

5.3. Peg stability

On the Chainlink mUSD/USD feed on Ethereum, the oracle mean is 0.9996 and the latest print is 0.9999. The largest downside deviation was to 0.9621, about 3.8% below peg, in mid-October 2025. Seven-day annualized volatility peaked near 1.5% in that window and now sits in a 0.1% to 0.4% range comparable to established stablecoins, even though mUSD is currently considered a very low liquidity asset and the feed is marked by Chainlink as being very high risk.


Source: LlamaRisk, July 15, 2026

5.4. On-chain and CEX liquidity

mUSD has no centralized-exchange listing, so all secondary liquidity is onchain, and MetaMask’s own buy page reflects an in-wallet, DEX-first design. On every chain the tradable liquidity concentrates in a single Uniswap v4 mUSD/USDC pool, with a long tail of dust pools carrying negligible balances and no volume.

Ethereum has one active venue. The Uniswap v4 USDC/mUSD pool is essentially all of the chain’s mUSD liquidity, and its high turnover (about 4x TVL in 24 hours) shows it is actively used.

Ethereum pool TVL 24h volume
Uniswap v4 USDC/mUSD $2.49M $10.24M

Monad is the deepest venue, again in one pool. The Uniswap v4 mUSD/USDC pool, created May 18, 2026, is the single deepest mUSD market anywhere and carries the Money Account flow.

Monad pool TVL 24h volume
Uniswap v4 mUSD/USDC $3.77M $4.10M
PancakeSwap v3 mUSD/USDC 0.01% $1.3K $0.9K

Linea holds a $1.0M Uniswap v4 pool, but that pool is nearly inactive (about $3.4K daily volume, heavily mUSD-skewed). Linea was the deepest chain at launch and has since thinned as supply migrated to Monad.

Linea pool TVL 24h volume
Uniswap v4 mUSD/USDC $1.01M $3.4K
EtherEx mUSD/USDC 0.01% $80.4K $29.1K

Translating pools into executable depth, live aggregator route quotes on July 15 absorb roughly $1.9M of mUSD into USDC within about 2% on Ethereum, roughly $2.0M to $2.3M on Monad, and only about $0.13M on Linea. The one-to-one Bridge redemption is permissioned and multi-day, so it does not backstop instant liquidation. For a lender or liquidator, the fast on-chain exit for mUSD is a single-pool DEX sale of low single-digit millions per chain.

6. The MetaMask Money Account on Monad

The Money Account is the reason of expected future mUSD adoption on Monad and the primary source of organic mUSD demand.

6.1. Product

Launched June 30, 2026, exclusively on Monad as the canonical chain, the Money Account is a self-custodial smart account built on account-abstraction architecture. Users hold their own keys, and MetaMask states it cannot access, freeze, or move a Money Account balance, with the ability to exit to an externally owned account at any time. Incoming USDC, USDT, DAI, and their Aave aToken equivalents auto-convert one-to-one into a single mUSD balance in one click with no fee. That balance earns up to 4% variable APY net of fees with no lockups, and links to a MetaMask Card on Mastercard with up to 3% back in mUSD. It is available globally except the UK and other restricted jurisdictions.

6.2. Yield stack

The APY is produced by a curated DeFi stack rather than by mUSD itself. Veda builds and operates the vault (a BoringVault architecture holding an mUSD liquidity sleeve for instant withdrawals), Steakhouse Financial curates the risk parameters and decides which lending markets the vault may touch, and capital routes into Morpho markets on Monad at launch (WETH/mUSD, wstETH/mUSD, WBTC/mUSD), with Aave to follow, as described in MetaMask’s launch announcement. Published TVL, depositor, and volume figures for the Money Account vault are not yet public. The observable proxies are Monad mUSD supply of about $9.56M and a Morpho WETH/mUSD market above $410K supplied mUSD in early July.

7. Deployment on Aave Monad

7.1. What supplying mUSD to Aave Monad means for Ethena

Ethena would deposit mUSD into the Aave Monad reserve and earn the supply rate plus any incentive, holding an aMonadMUSD position. This mirrors Ethena’s existing Aave supply of about $528M across different markets and assets. The relevant risk is liquidity recoverability: Ethena’s mUSD allocation would be the supermajority of the reserve’s supply on Monad. This means that in practice, Ethena’s ability to pull mUSD backing quickly is to be bounded by Aave Monad available liquidity, external supply in the mUSD reserve on Monad and utilization. It is notable that the diversification in allocations has historically allowed Ethena to remain fully liquid, without immobilizing the liquid backing buffer.


Source: Ethena Transparency Dashboard, July 15, 2026

8. Suitability as a USDe backing asset

8.1. Role in USDe architecture

As of mid-2026 USDe supply is roughly $4.0B, down about 70% from a peak above $14B, and its backing has rotated toward liquid stablecoins and DeFi lending, with delta-hedged crypto now a minority of reserves. A par-redeemable, T-bill-backed stablecoin like mUSD belongs in that liquid-stables allocation, which is currently the largest backing category, so adding it extends what Ethena is already doing rather than introducing a new kind of exposure. The yield is earned by supplying the stablecoin into Aave and Morpho as well as incentives, which is the proposed structure.

8.2. Comparison with approved backing stablecoins

mUSD’s reserve and yield structure matches assets Ethena already holds or has assessed, it differs mainly on maturity, size, and liquidity.

mUSD USDtb USDG
Issuer Bridge (Stripe) / M0 Ethena / Securitize / Anchorage Paxos
Reserve Cash + short-term USTs ~90% BlackRock BUIDL Cash, short-term USTs, MMF, repo
Primary regulation FinCEN MSB, pending trust charter, GENIUS-ready Securitize / BUIDL framework MAS + MiCA + GENIUS
Named reserve attestor None yet (Chronicle on-chain) BlackRock / Securitize KPMG
Redemption Permissioned, KYC, multi-day Permissioned, KYC Permissioned
Secondary liquidity Thin, ~$2-4M/chain, no CEX ~10M DEX, no CEX ~$10M Curve + CEX
Supply ~$32M ~$1.0B ~$1.7B

On the reserve itself, mUSD is not a riskier kind of asset than what Ethena backs USDe with today. It is a smaller and less liquid instance of the same conservative T-bill-backed profile.

8.3. Risk factors

  1. Liquidity and exit depth. About $2M to $2.5M of near-par on-chain depth per chain and no CEX market. This is the binding constraint, so any allocation should be sized to stress-case exit throughput, not to Aave’s cap.
  2. Redemption is permissioned and multi-day. The one-to-one issuer redemption is real but KYC-gated at the ramp without instant capacity, so it does not substitute for onchain liquidity.
  3. Small size and concentration. About $32M total supply concentrated on Monad and dependent on a single product, the Money Account, for organic demand; a shared Steakhouse curator and shared Monad venues create a correlated dependency with Ethena’s own deployment.

9. Recommendation

LlamaRisk supports onboarding mUSD as a USDe backing asset and supports deploying that mUSD into the Aave v3.7 instance on Monad. This deployment works the same way as Ethena’s existing Aave and Morpho stablecoin supply: the reserve is conservative and consistent with Ethena’s approved backing set, Ethena already earns yield by supplying stablecoins into these markets, and the Monad venue is live and already lists mUSD.

Scale-up should be conditioned on monitoring across the following dimensions:

  • Redemption performance at par and under stress, including realized settlement timing on a real Ethena redemption.
  • Publication of recurring third-party reserve attestations from a recognized firm.
  • Durability of Money Account demand and the health of the shared Steakhouse-curated Monad lending markets.

On that basis, mUSD is a suitable, if small and early-stage, addition to the liquid-stables portion of USDe’s backing.

Disclaimer

This review was independently prepared by LlamaRisk, a risk service provider serving in Ethena Risk Committee. LlamaRisk is also the sole risk service provider of Aave. The information provided should not be construed as legal, financial, tax, or professional advice.