⚠ V1 pre-launch — design and Aiken implementation both complete. External audit target Q2-Q3 2027. App not yet available.

Auto-Yield
Stablecoin Vault
on Cardano

A Cardano DeFi public-goods reference implementation — a non-custodial USDCx auto-yield vault built by a Cardano self-custody user who couldn't find one on-chain and decided to ship it. Not optimized for commercial scale; Apache 2.0, fork-welcome. Core user protections are contract invariants, not operational promises. V1 design complete; public launch gated on external audit (target Q2-Q3 2027).

22
Artefacts
235
Tests / 730 Checks
15
Immutable Fields
$100K
Pre-Audit Cap

Phase 1 expected TVL $500–$25K (speculative estimate, not a forecast). Pre-audit cap is a risk envelope, not a sales target. Hard-capped at 4.5% performance fee; 29-field VaultDatum with 15 immutable anchors; 17 logic validators + 4 NFT mint policies + 1 DEX adapter under Plutus V3 16 KB ceiling.

Why This Team Is Building It

The founder is a member of the target user segment — a Cardano self-custody holder who couldn't find a non-custodial USDCx auto-yield vault and decided to build one.

The origin — a Cardano user's personal problem

OptiVaults's founder is a Cardano self-custody user: ADA holder + native staker (delegated to a stake pool) since 2025, small-BTC holder since 2025, participated in Midnight's NIGHT redeem (those NIGHT tokens later ended up paired with ADA on Minswap V2 — a natural continuation, not a planned portfolio construction), Minswap V2 ADA/NIGHT LP since 2026, and a current USDCx holder (USDCx launched on Cardano via Circle's xReserve in February 2026). V1 did not start as a market opportunity — it started as a specific personal problem: USDCx was in the founder's wallet and the realistic options were (a) send to a CEX earn product (rejected on self-custody principle), (b) supply directly on Liqwid's USDCx market for ~0.5-2% APY (inflation-negative), or (c) swap to DJED and supply on Liqwid for ~11.8% — but (c) still requires a 2-TX entry per market (swap USDCx → DJED, then Liqwid supply), parallel management if you want DJED + USDM + USDCx diversification, and a 2-TX recall + swap-back round-trip per full or partial withdraw. The qToken sits in the wallet between actions and its rate accrues automatically — no per-cycle chore. On a $200 position the entry + exit round-trip gas is ~1-2% of principal. No "deposit USDCx and let it auto-compound across DJED + USDM with a single-TX exit" non-custodial vault existed on Cardano. So the founder built one.

Why Cardano, why now

Aave, Compound, and Yearn shipped conceptually similar vault-aggregator patterns on Ethereum years ago. V1 is being built on Cardano in 2026 because the preconditions only recently aligned: USDCx launched via Circle's xReserve in February 2026, Liqwid's stablecoin markets reached supply-depth maturity in 2024-2025, the Cardano SPO operator community reached breadth needed for credibly-independent governance signers, and Aiken + PlutusV3 tooling stabilized to v1.1.x. V1 is not the first auto-yield vault in DeFi — it is the first version whose architectural preconditions are actually satisfied on Cardano.

Builder contribution — why Apache 2.0, and an honest view of V2

OptiVaults is not trying to become Cardano's largest vault. The explicit goal: put a complete, audited, Apache-2.0-licensed reference implementation into the Cardano open-source commons — so that even if OptiVaults stays small, other teams can fork V1 and specialise it. V2+ would evolve this same idea toward a multi-protocol, multi-strategy asset-allocation layer as the Cardano DeFi landscape deepens — a direction, not a committed roadmap. If that evolution does not happen, V1 remains what it offers today: automatic compounding + dual-issuer stablecoin allocation + self-serve exit + audited-and-capped fee structure, costing 4.5% of realised yield. We'd encourage depositors to evaluate V1 on those concrete properties rather than on any future V2 features.

How It Works

Three simple steps to start earning yield on your stablecoins.

1

Deposit

Connect your Cardano wallet and deposit stablecoins into the vault. You receive vUSDCx share tokens representing your proportional claim.

2

Earn

The automated keeper deploys capital across DeFi protocols, compounds yields, and rebalances — all validated on-chain.

3

Withdraw

Burn your vUSDCx shares anytime to receive your original deposit plus accumulated yield. No lock-ups, no permission required.

Yield Strategy

A diversified, risk-managed approach to stablecoin yield on Cardano.

💳
30% USDCx Idle Buffer

30% of deposits stay as idle USDCx at the vault for instant withdrawals (0% yield at V1 launch by design — buffer must stay liquid). Governance-adjustable via UpdateStrategy 7-day timelock if a yield-bearing home becomes viable.

📈
45% DJED + 25% USDM

45% DJED + 25% USDM supplied to Liqwid Finance stablecoin markets (launch target). Depositors bear dual-issuer allocation: Cardano Foundation/COTI for DJED, Mehen for USDM — within single-ecosystem Cardano exposure.

⚙️
Dynamic Allocation

Risk engine adjusts weights based on real-time APY, pool depth, and depeg signals. Governance can override via 7–21 day timelock (action-dependent).

Liqwid Yield Cycle (no DEX swap needed)

1. Recall from Liqwid (burn qTokens, receive underlying + interest) → 2. Compound (on-chain yield detection, fee deduction) → 3. Re-deploy to Liqwid. Capital-efficient — yield realized directly in the recalled token.

Expected Returns

Illustrative projections after the 4.5% performance fee. Actual returns depend on market conditions.

Scenario Gross APY Net APY
Conservative ~2% ~1.9%
Moderate ~5% ~4.8%
Optimistic ~8% ~7.6%

APY is variable and not guaranteed. Past performance does not predict future results.

Transparent Fees

Simple, on-chain enforced fee structure. No hidden charges.

4.5%
Performance Fee

Only charged on yield earned — never on your principal. Enforced on-chain at compound time.

Try it — enter your deposit:
USDCx
Assuming 6% APY over 1 year:
600.00
Gross Yield
27.00
Fee (4.5%)
573.00
Your Net Yield
5.73%
Effective APY
0.1%
Direct Withdraw Fee

Instant withdrawal from the 30% idle buffer. The 0.1% fee stays in the vault — it raises the share price for all remaining holders (deferred yield).

0.5 / 1 ADA
Queue Deposit / Withdraw

Batcher fee covering the keeper's Conway ref-script TX cost. ~2 ADA network deposit returned on cancel. Direct deposit (wallet→vault) has no OptiVaults fee beyond the ~1 ADA network TX fee.

How Yield Distribution Works

OptiVaults uses a share-based model (similar to ERC-4626) where all depositors earn yield proportionally.

1. Deposit USDCx → Receive vUSDCx

You receive vUSDCx share tokens proportional to the current vault value. Share price starts near 1:1 and rises as yield compounds.

2. Yield Compounds → Share Price Rises

When the keeper harvests yield, total vault value increases but total shares stay the same. This raises the share price equally for all holders.

3. Fair for Everyone

New depositors get fewer shares at the higher price — they don't dilute existing holders. Early depositors benefit from lower entry price. Everyone earns proportionally from their deposit moment.

4. Withdraw Anytime

Burn vUSDCx to receive your proportional USDCx back (principal + earned yield). 0.1% fee on direct withdrawals (free via queue). The fee stays in the vault, benefiting remaining holders.

Security First

Built from day one with security as the top priority. Every operation validated on-chain.

15 Immutable Fields

Admin keys, fees, policies, and registry auth are locked at deployment. No operation can alter them.

Exact-Value Validation

Deposit and compound outputs are validated with exact equality — no room for value extraction.

Role Separation

Admin, Keeper, and User roles have distinct keys and capabilities. No single key controls everything.

Flash Loan Resistant

Direct withdrawals always charge a fee. Batched withdrawals require queuing. Both prevent same-block exploits.

Inflation Attack Protection

Share multiplier and minimum deposit work together to block ERC-4626 style inflation attacks.

Slippage Protection

Token-specific minimum receive amounts enforced on-chain. Users are protected against sandwich attacks.

Multi-Round Internal Audit

Multiple rounds of internal audit review completed during pre-launch; V1-specific internal audit coverage + external audit (target Q2-Q3 2027) gate the public launch.

Open Source Contracts

Smart contract code is publicly available for independent verification and community auditing.

Depeg Guard

Automatic stablecoin price monitoring with staged withdrawal on significant deviation. 30% buffer provides additional safety margin.

7-Day Emergency Escape

If the keeper goes offline for 7+ days, any user can withdraw directly from the smart contract — no admin approval needed. Early withdrawal fee is also waived.

Admin Fallback

After 7 days of keeper inactivity, admin can recall deployed funds. General deploy remains keeper-only. Only stranded non-deposit tokens (e.g. DJED/USDM) can be swapped by admin via AdminDeployNonDeposit — subject to a 21-day Registry timelock. Admin cannot touch USDCx or ADA.

On-Chain Governance

Transparent, time-locked, multi-signature governance with user exit guarantees.

3-of-3 Unanimity at Launch

All governance actions require unanimous signer approval at V1 launch (1 founder + 2 independent Cardano SPOs). Timelocks: 7–21 days depending on action (EmergencyWithdraw 0d / UpdateSlippagePolicy 48h / UpdateStrategy + TreasurySpend 7d / UpdateFee + Registry + KeeperAuth 14d / UpdateFeeSplit 21d). Any single signer can 1-of-n cancel during timelock.

7-Day Timelock

All non-emergency governance actions require a minimum 7-day waiting period. Users can exit before any change takes effect.

15 Immutable Fields

Critical parameters (fee collector, minting policy, keeper key) are enforced on-chain and cannot be changed after deployment.

7-Day User Exit

If keeper goes offline for 7+ days, any user can force-withdraw directly from the smart contract. No governance approval needed.

Self-Serve Recovery

Your funds are always recoverable — even if our infrastructure goes offline permanently.

🔧
Withdraw CLI

Open-source npm package. Build and sign withdrawal transactions locally with your own Blockfrost key. No backend needed.

🌐
Emergency Page

Self-contained HTML file (works offline). Connect wallet, enter Blockfrost key, withdraw. Save it now as your insurance policy.

👥
Community Frontend

All contracts are open source (Apache 2.0). Anyone can build an alternative frontend with the same contract addresses.

Risks & Pre-Audit Disclosures

OptiVaults is in a pre-audit deployment phase. Principal is at risk. Review all risks before depositing.

⚠️ Pre-audit TVL cap: USD 100,000

Total value locked is operator-capped at USD 100,000 equivalent until the professional third-party audit (scheduled Q2-Q3 2027) completes. The cap is enforced at the frontend and by keeper-side monitoring. Do not deposit more than you can afford to lose.

Smart Contract Risk

The contracts have passed multiple rounds of internal audit review (findings resolved), but have NOT yet been reviewed by an independent third-party auditor. A professional audit is scheduled for Q2-Q3 2027. Residual bugs cannot be ruled out.

Stablecoin Depeg Risk

USDCx (Circle via xReserve), DJED (COTI algorithmic), and USDM (Mehen fiat-backed) can lose their USD peg. Automatic depeg monitoring triggers phased recall at >5% deviation, but cannot undo realised losses.

Liqwid Protocol Risk

Deployed capital earns yield via Liqwid Finance lending markets. Liqwid smart-contract bugs, bad-debt events, or market-wide liquidity crises directly impact vault NAV. RecallFromLiqwid may return less than supplied during protocol distress.

Keeper Infrastructure Risk

The keeper runs as dual-instance shared-key automation. Key compromise cannot drain principal (on-chain contracts prevent this) but can stall compounding and batch processing. 7-day keeper-inactivity fallback lets any user withdraw directly.

Governance Risk

V1 launch governance is 3-of-3 MultisigGov (unanimity required): 1 founder + 2 independent Cardano SPOs. Action-specific timelocks 7–21 days (UpdateFeeSplit longest at 21d, self-dealing protection) + 1-of-n cancel veto by any signer during the timelock window. Full 3-of-3 collusion could signal malicious actions (fee/strategy changes), but 15 immutable datum fields + hard-capped 4.5% fee + emergency withdraw bound the blast radius.

On-Chain Execution Risk

Cardano eUTXO congestion, DEX slippage on Minswap V2 routes, or temporary Blockfrost/Ogmios indexing delays can postpone — not prevent — deposit / withdraw / compound. Funds remain on-chain throughout.

Full risk disclosure: Whitepaper §5 · Protocol docs (LLM-friendly) · Vulnerability reporting

Why OptiVaults?

A smarter way to earn yield on your stablecoins without the complexity of manual DeFi farming.

Non-Custodial

Your funds are held by on-chain smart contracts, not by us. Withdraw anytime with your wallet — no permission needed.

Auto-Compounding

Yields are harvested and reinvested automatically. Your share value grows over time without any manual action.

On-Chain Security

15 immutable fields, exact-value validation, and role separation. All critical operations enforced by Aiken PlutusV3 validators.

Optimized Allocation

A risk-aware strategy engine allocates across multiple DeFi protocols to maximize risk-adjusted returns.

Mobile-First

Works with any CIP-30 wallet — Vespr, Eternl, Nami, Lace. Server-side TX building means no WASM on your device.

Open Source

Smart contracts are open source under Apache 2.0. Verify the code, audit the logic, trust the math.

For Builders, Auditors, Researchers, Forkers

V1 is a Cardano DeFi public-goods reference implementation. Direct entry points for technical readers.

Auditors

Researchers

Forkers

Review history: V1 docs went through 9 external-reviewer feedback cycles + multi-round internal audit coverage A–F summarised in docs/audit-scope.md. Active project log mirrors are in the private development tree.

Frequently Asked Questions

Everything you need to know about OptiVaults.

USDCx is a USD-pegged stablecoin on Cardano issued by Circle via its xReserve cross-chain reserve mechanism (launched February 2026). Backed 1:1 by Circle's USDC reserves with monthly Deloitte attestations. Uses Cardano native asset format — directly consumable on-chain without a wrapping contract. It is the primary deposit asset for OptiVaults. You can acquire USDCx on Minswap V2, SundaeSwap, or via Circle's xReserve bridge.
vUSDCx is a share token representing your proportional ownership of the vault. As yield compounds, the share price rises — so your vUSDCx becomes worth more USDCx over time.
Yield is shared proportionally among all vUSDCx holders. When the keeper compounds yield, total_deposited increases but total_shares stays the same — raising the share price for everyone equally.
Performance fee: 4.5% on yield earned only (never on principal). Early withdrawal fee: 0.1% on direct withdrawals (waived for queue withdrawals and emergency mode). Network TX fee: ~1 ADA per Cardano transaction. Queue deposit: 0.5 ADA batcher fee (~2 ADA refunded on cancel). Queue withdraw: 1 ADA batcher fee (~1.5 ADA refunded on cancel). No OptiVaults fee on direct deposits.
No. Admin and Keeper use separate keys with strictly separated permissions. 15 immutable fields are enforced on-chain. Admin cannot deploy funds (keeper-only) and all operations preserve token balances verified on-chain. All operations verified by comprehensive test suite.
If the keeper hasn't compounded for 7+ days, emergency mode activates. Any vUSDCx holder can withdraw directly from the smart contract — no admin approval needed. The early withdrawal fee is also waived.
V1 contracts will be audited internally across multiple rounds and by an independent third-party firm (target Q2-Q3 2027) before full public launch. All contract code is open source (Apache 2.0) and verifiable on GitHub.
Any CIP-30 compatible Cardano wallet: Eternl, Nami, Lace, Flint, Yoroi, and others. No custom wallet integration needed.
Yes. Principal is at risk. Main scenarios: (1) USDCx / DJED / USDM stablecoin depeg; (2) Liqwid protocol failure or bad-debt; (3) smart-contract bug not caught by internal audit (Q2-Q3 2027 third-party audit pending); (4) keeper infrastructure compromise (cannot drain principal but can stall compound / batch). The 30% idle buffer, 4.5% fee hard cap, depeg monitoring, m-of-n timelock, and 7-day emergency-withdraw escape reduce but do not eliminate these risks. See the Risks section above and whitepaper §5. Start with a small amount.

Roadmap

From smart contract development to a fully decentralized yield protocol.

Phase 1 — Smart Contracts
Aiken PlutusV3 vault with multiple redeemers, extensive on-chain property-based testing, multi-round internal audit review
Phase 2 — Keeper Infrastructure
Automated compound, rebalance, batch processing, risk engine, multi-protocol integration
Phase 3 — Frontend & API
React SPA with CIP-30 wallet support, server-side TX building, JWT auth, real-time WebSocket updates
Phase 4 — Preprod Verification
Multiple rounds of on-chain E2E testing, all 14 redeemers verified across full deposit-to-withdraw lifecycle
Phase 5 — Internal Verification (Sunset Complete)
Prior internal-verification vaults (V5 through V10) operated on Cardano mainnet under founder-only test access for ~18 months: MultisigGov m-of-n governance, Withdraw-Zero forwarding pattern, one-shot Vault Identity NFT compile-time anchor, per-market Liqwid position tracking, multi-round security audit review with findings addressed. Design patterns validated under real market conditions — then drained or in active sunset/recovery. V1 is a fresh Aiken rewrite and fresh mainnet deploy, not a patched legacy vault: new contract hashes, new audit scope, new deploy ceremony. V1 inherits the validated design patterns but no legacy state.
Phase 6 — V1 Design Complete
V1 design + Aiken implementation complete. 17 logic validators + 4 NFT mint policies + 1 DEX adapter = 22 compiled artefacts, 506 unit + property tests / 1,496 randomized checks pass. 3-way Compound fee split (V1 launch 40% keeper / 0% gov / 60% treasury), weekly keeper round-robin rotation, quarterly gov-signer compensation with X3 hybrid qualification, soul-bound Gov Signer NFT, UpdateFeeSplit 21-day timelock, UpdateSlippagePolicy 48-hour timelock, Phase 1 three-layer governance safety (Layer 1/2/3 freeze + swap-out + 90-day CommunitySunset), 29-field VaultDatum (15 immutable), on-chain treasury with 4-category budget. Apache 2.0 license across the tree. Public whitepaper + product overview + security model + migration plan + audit scope all published.
Phase 7 — V1 Aiken Implementation & Internal Audit
Aiken PlutusV3 implementation complete (17 logic + 4 NFT + 1 adapter); property-based fuzz suite live (10 properties × 100 iterations per `aiken check`); V1 Preprod deploy ceremony LIVE with the numbered E2E test suite chain-verified end-to-end; V1-specific internal audit coverage areas A-F (treasury, keeper_stake_script, integration flows, off-chain runtime, deploy pipeline) covered through multi-round internal review.
Phase 8 — External Audit & V1 Mainnet Launch
Third-party security audit (target Q2-Q3 2027), audit findings resolved, V1 mainnet deploy ceremony, migration window opens for pre-launch test participants, public launch with 100K USDCx TVL cap pending post-audit lift.
Phase 9 — Governance Scaling & Keeper Permissionless
Staged governance rotation (3→5→7 signers), Phase 3 community signer election (mechanism TBD — not vUSDCx-weighted), keeper PermissionlessWithBond mode when TVL economics support external operators (~$5-10M).

Follow V1 Development

App launches after external audit (target Q2-Q3 2027). Read the design, review the code, join the community.

View Source View Roadmap
Top How it Works Features Yield Fees Security FAQ Roadmap