Euler V2

Euler V2 is a modular, permissionless lending protocol built around independent ERC‑4626 vaults (Euler Vault Kit, EVK) and the Ethereum Vault Connector (EVC), relaunched as Euler v2 after the original Euler exploit and recovery. Identification

• Name: Euler V2 (Euler Finance v2)
• Website: Euler Finance main site and docs subdomain (exact URLs not provided per instructions).
• Docs: Euler V2 documentation covering introduction, lite paper, EVK/EVC, developers’ guide, and llms index.
• Category: Lending / money‑market protocol using ERC‑4626 credit vaults.
• Launch / relaunch date: v2 described as a redesign following the March 2023 exploit; independent analysis and Euler materials place the v2 relaunch around September 2024.
• Supported chains (for this slug): Arbitrum, Avalanche, BSC (BNB Chain), Base, Ethereum, Hyperliquid L1, Linea, Monad, Plasma, Unichain, consistent across independent analytics and protocol‑level descriptions.
• Native / governance token: EUL is the protocol token referenced in docs and risk/governance materials. Main contract addresses & verification
• Docs explicitly state a Treasury address on Ethereum: 0xC7C5aFDB61e08BE3e2FB09098412b5706EB5c550.
• Public documentation references EVK vault contracts and the EVC but does not enumerate canonical “main” addresses per chain in the excerpts retrieved.
• Explorer verification status (e.g., contract verification for EVK/EVC on each chain) is Not verifiable as of 2026‑08‑29 under current data, because we lack direct explorer lookups in this run.
• Full on‑chain contract lists and deployment addresses per chain are likewise Not verifiable as of 2026‑08‑29. Fork lineage, design origin, and modifications
• Euler v2 is presented as a redesigned successor to the original Euler lending protocol, which suffered a major exploit in March 2023, later fully recovered, and then relaunched as v2 with a “completely redesigned modular architecture.”
• The architecture centers on custom EVK credit vaults (ERC‑4626 with borrowing) plus EVC for cross‑vault collateral, moving away from the monolithic pool model of v1.
• There is no evidence in retrieved sources that Euler V2 is a fork of another protocol; it is described as a proprietary redesign rather than, for example, an Aave or Compound fork.
• Changes versus Euler v1 include modular vaults, EVC interoperability layer, and EulerEarn meta‑vault aggregation.
• Independent analytics state that Euler V2 “has been independently audited, with no recorded security incidents to date,” but specific auditor names and scope are not shown in the excerpts, so those details are Not verifiable as of 2026‑08‑29.
• No records in the retrieved data indicate malicious‑modification history in Euler‑related forks; such incidents, if any, are Not verifiable as of 2026‑08‑29.

Euler V2 appears to be a real, functional product rather than a pure landing page: the docs explicitly direct users to the official app, describe live flows for supply, borrow, repay, withdraw, and Multiply, and the user guide says direct deposits live in the user’s main wallet address while withdrawals depend on available liquidity. The documentation also exposes developer material and a data-querying section, which is a strong maturity signal, and it states that an API is currently in the works for integrators. Open API: Not clearly public for the product core as of 2026-08-29. The docs say an API is “in the works,” while separate third-party documentation references a GraphQL endpoint/subgraph and a draft OpenAPI file, but those do not by themselves prove a stable, official public API surface for the protocol UI or end users. UI/UX quality: The source set shows a maintained docs site with structured guides and developer references, but there is no direct evidence here of broken links, fake metrics, or template-site signs. Not verifiable as of 2026-08-29. Live deposits/withdrawals: The docs describe them as available, but without on-chain verification in this run, actual live execution and chain-by-chain availability are Not verifiable as of 2026-08-29. For the same reason, chain-specific maturity breakdown across Arbitrum, Avalanche, BSC, Base, Ethereum, Hyperliquid L1, Linea, Monad, Plasma, and Unichain is Not verifiable as of 2026-08-29.

Across Euler V2, public sources report 29–31+ audits by ~12–13 independent firms plus large‑scale audit competitions. Media and protocol communications agree that these engagements collectively identified two critical and five high‑severity issues, all reported as fixed before launch. Subsequent public bug bounty and Cantina audit‑contest activities reportedly found no remaining critical, high or medium‑severity issues; only low‑severity bugs were rewarded. Exact mapping of each firm’s report to specific modules, chains, and deployments remains fragmented and not verifiable as of 2026-08-30 at a bytecode‑match level.

yAudit performed a comprehensive, independent audit of the Euler V2 EVC, EVK and EPO codebases over a 30‑day period. The engagement covered core vault connector logic, vault kit, and price oracle modules for the v2 architecture. Detailed findings (including severity and fix status) are in the PDF report; public secondary sources only state that issues were resolved, with no disclosed remaining ≥medium‑severity items. Not verifiable as of 2026-08-30 whether this report’s scope covers all currently deployed chain variants or later upgrades.

Euler V2 has an active Cantina bug bounty program. The live program page shows a start date of 21 Aug 2024, while Euler’s own docs still describe an active live Cantina bounty in 2026. Parameters and scope: the program covers the vaults the Euler app marks as known on each network and the contracts those vaults directly rely on; it also covers core V2 components including Euler Vault Kit (EVK), Ethereum Vault Connector (EVC), Euler Price Oracle (EPO), supporting components, EulerEarn, EulerSwap, and the web interface. Rewards on the current Cantina program page are listed as $25,000 for critical, $5,000 for high, and $1,000 for medium web-interface findings, while the page also states a total reward pool of 7,500,000 USDC + rEUL + USUAL and notes a $20 deposit requirement. Euler’s docs separately state rewards can reach up to $7.5M for critical vulnerabilities. Results: the current Cantina bounty page lists findings submitted, but the retrieved snapshot does not provide a published tally of resolved vulnerabilities or payouts. Euler’s own older post-launch materials say a prior $1.25M post-audit bug bounty found no medium-or-higher severity issues and paid $200,000 for low-severity findings. A separate 2022 Immunefi-era bounty existed for Euler v1, but that is not the current Euler V2 program. Not verifiable as of 2026-08-29: a complete, independently published results breakdown for the current Euler V2 Cantina program.

Euler v2’s risk profile is dominated by its oracle and integration dependencies, plus the quality of assets it lists (stablecoins, LSTs, restaking/RWA), rather than direct custodial or CEX exposure. On‑chain verification is Not verifiable as of [2026-08-29]. ### 1. External protocol integrations

• Base layer: Euler v2 is an overcollateralized lending protocol using isolated markets (Euler Vaults); design aims to contain contagion between assets but still relies on correct collateral pricing and liquid secondary markets.
• DEX/liquidity dependence: Liquidation, collateral valuation and slippage assumptions depend on spot liquidity on major DEXs (Uniswap, Curve, etc.). Thin liquidity assets listed in Euler Vaults increase liquidation failure and price impact risk. ### 2. Oracle & price manipulation risk
• Euler v1’s historical design used Uniswap TWAP oracles and external price feeds, and its 2023 exploit involved complex MEV/manipulation of collateral and borrowing positions. That history is a material risk precedent for v2 design and monitoring.
• v2 still depends on external price oracles (likely Chainlink for major assets plus DEX‑based mechanisms for long‑tail tokens); any oracle failure, lag, or manipulation at times of low liquidity could cause:
• underpriced collateral → insolvent markets
• overpriced collateral → bad debt/loss to lenders
• Concentrated liquidity pools or single‑venue pricing deepen manipulation risk (flash loan attacks, swing pricing). ### 3. Bridges & cross‑chain
• Euler v2 is multi‑chain (Arbitrum, Base, etc.). Cross‑chain asset flows use third‑party bridges (e.g., canonical rollup bridges, popular bridging protocols). Bridge failure, censorship or exploit can strand collateral or cause chain‑specific depegs. Specific bridge sets per chain: Not verifiable as of [2026-08-29]. ### 4. Custodians, CEX/MM exposure
• As a non‑custodial protocol, Euler does not hold assets off‑chain; user funds are in smart contracts. Direct custodian/CEX dependence is limited to:
• off‑chain liquidity providers/market makers whose behavior affects on‑chain prices
• centralized stablecoin issuers’ redemption policies. ### 5. Asset‑level (stablecoin, LST, RWA, restaking) risk
• Stablecoins: Exposure to USDC/USDT and other fiat‑backed coins introduces issuer/regulatory risk, blacklisting, and depeg scenarios. Euler markets could become undercollateralized if a major stablecoin depegs.
• LSTs & restaking: If Euler v2 lists LSTs or restaked/LRT assets, their protocol‑level slashing, liquidity and oracle risks compound lending risk. Exact LST/restaking set per chain: Not verifiable as of [2026-08-29].
• RWA/SPVs: Any RWA collateral would depend on off‑chain SPVs, legal enforceability and valuation transparency; current RWA use in Euler v2: Not verifiable as of [2026-08-29]. ### 6. Key failure scenarios
• Major oracle failure or manipulation on a thin‑liquidity asset.
• Stablecoin/LST depeg on a heavily used collateral.
• Bridge exploit affecting a chain where Euler holds large positions.
• Low‑liquidity isolated market becoming insolvent, with shortfall socialized through protocol‑level mechanisms. Sources: Euler Docs & risk/disclosure materials for Euler Vaults and multi‑chain deployment.

Euler V2 organizes custody as a non-custodial, modular vault system: users deposit assets into separately deployed ERC-4626 vaults rather than into one shared pool, and each vault is the basic unit of custody and risk. The Euler Vault Kit (EVK) lets anyone deploy configurable vaults permissionlessly, while the Ethereum Vault Connector (EVC) lets selected vaults recognize positions in other vaults as collateral and compute solvency across connected vaults without merging them into a single pool.

Since the Euler v2 relaunch, no publicly reported smart-contract exploit, oracle/manipulation event, or loss incident has been attributed to Euler v2 on any of the listed chains (Arbitrum, Avalanche, BSC, Base, Ethereum, Hyperliquid L1, Linea, Monad, Plasma, Unichain) as of 2026‑08‑29. Any on-chain confirmation is Not verifiable as of 2026‑08‑29. The major historical incident concerns Euler v1 (Ethereum):

• Date: Attack executed 2023‑03‑13; negotiations and returns through 2023‑04‑04.
• Cause: A missing health check after a donateToReserves call allowed a flash‑loan driven self‑liquidation pattern that generated bad debt absorbed by the attacker, exploiting the liquidation discount mechanics.
• Loss magnitude: ~$197m drained across DAI, WBTC, stETH, USDC and others.
• Affected: Euler v1 lenders/borrowers on Ethereum; the protocol’s solvency was heavily impacted.
• Response: Euler engaged the attacker via on‑chain messages, offered a bounty and negotiated return of funds.
• Reimbursement: The attacker eventually returned essentially all funds (~$240m including appreciation), enabling user restitution.
• Fix: Euler v1 was effectively sunset; v2 was rebuilt with modular vault architecture and extensive security hardening. Euler v2 security posture / bug bounty
• Audits: V2 codebase has undergone 29–31+ external audits by 12–16 security firms, plus a large Cantina audit competition; over $4m spent on security for v2.
• Bug bounty: Live Cantina bounty and broader program with rewards up to $7.5m for critical vulnerabilities, focused on the Euler Vault Kit, Vault Connector, Oracle, and web front‑end. A dedicated v2 Cantina competition with $1.25m rewards found no medium‑or‑higher‑severity issues; ~$200k was paid for low‑severity findings. Key management / custody / key‑person risk
• Detailed on‑chain role distribution and multisig setups for v2 are Not verifiable as of 2026‑08‑29.
• Public materials emphasize modular “vault” architecture, heavy external audits, and reliance on formalized emergency frameworks (e.g., SEAL Safe Harbor) for whitehat intervention, which mitigates but does not eliminate key‑person and operational risk.
• Given the v1 exploit history and the centralized development by Euler Labs, governance and upgrade authority for v2 should be treated as non‑trivial key‑management and key‑person risk pending direct verification of multisig composition and timelock parameters (Not verifiable as of 2026‑08‑29).

Euler V2’s key management is organized around *on-chain roles and vault-level control rather than a single protocol master key*. The protocol’s core coordination layer, the Ethereum Vault Connector (EVC), is described as an immutable, open-source contract with “no upgrade keys,” and it provides unified authentication plus up to 256 isolated sub-accounts per wallet so users can separate positions without extra wallet keys. For user-side delegation, the EVC supports *operator* functionality, where a smart contract or EOA can be delegated to act on a user’s behalf. At the market level, key management depends on the vault type. Euler documents describe *permissionless* EVK vaults that anyone can deploy, and *governed* vaults where a DAO, risk manager, or individual can control parameters such as risk settings and interest rates; some vaults are also described as *unmanaged*, meaning no third party can change parameters after launch. This means administrative power is not centralized across the whole protocol, but split between immutable shared infrastructure and per-vault governance or autonomy. One important caveat is that protocol docs and third-party commentary agree on the modular, key-minimized design, but a chain-by-chain review of deployed admin keys and timelocks is *Not verifiable as of 2026-08-29* from the provided sources alone.

Euler v2 is a new, multi‑chain deployment; on-chain inspection via Dune is Not verifiable as of 2026‑08‑29. Below is based on public docs, audits and v1 precedent, so treat as preliminary. 1. Core contracts & verification

• Euler v2 is live or planned on Ethereum, Arbitrum, Base and other L2s, but precise core contract addresses, verification status and proxies per chain are Not verifiable as of 2026‑08‑29.
• v1 used an upgradable proxy pattern with separate modules for lending, risk, and liquidation, suggesting v2 likely remains modular and upgradeable, not a single immutable contract. 2. Upgradeability & admin architecture
• Euler v1 used a governance contract controlling module upgrades and risk parameters, with an emergency admin capable of pausing markets.
• For v2, similar governance and admin control is described in docs (protocol governance controlling upgrades and parameters), but exact proxy admin type, timelock contracts and delays per chain are Not verifiable as of 2026‑08‑29.
• No reliable evidence that v2 has fully renounced admin or owner roles on any chain. 3. Permissions & emergency controls Typical high‑impact functions (inferred from Euler design and DeFi norms; not on‑chain verified):
• Pause / circuit breaker: ability to pause borrowing, liquidations, or entire markets in emergencies.
• Parameter changes: collateral factors, interest‑rate models, reserve factors, listing/delisting markets.
• Upgrades: swapping implementations of core modules (lending, liquidation, risk engine).
• Fees & treasury: adjusting protocol fee rates and fee recipients.
• Oracle/strategy hooks: changing price oracle sources, which materially affects liquidation and solvency. 4. User exit & worst‑case key scenarios
• Design intent (from v1 and docs) is that users can withdraw when markets are solvent and not paused; however, whether withdrawal is always available without admin cooperation on v2 across chains is Not verifiable as of 2026‑08‑29.
• If governance/admin keys are compromised, attacker could:
• Upgrade implementations to malicious code (steal collateral or deposits).
• Change oracles to trigger forced liquidations.
• Redirect protocol fees / reserves.
• Perma‑pause markets, effectively freezing user funds.
• This implies rug/freeze risk is non‑zero until full decentralization (timelocked, multi‑sig or DAO control with robust processes) is confirmed. 5. Architecture map (conceptual)
• Proxy layer: chain‑specific proxies for lending pools and modules.
• Implementation layer: interest‑rate, risk, liquidation, oracle, and collateral logic.
• Governance/admin layer: multisig/DAO contracts with upgrade, pause and parameter powers.
• Oracle layer: external price feeds integrated via dedicated contracts. Due to lack of reliable on‑chain and audit data for Euler v2, many details are Not verifiable as of 2026‑08‑29 and should be treated as open risk pending direct contract inspection and formal audits.

Euler V2 appears to be a real operating team, not an anon front, but the best-supported picture is mixed: the project publicly identifies three co-founders — Michael Bentley, Jack Prior, and Doug Hoyte — and Bentley has publicly described Euler as founded in 2020. Bentley is presented in independent coverage as a former Oxford researcher/post-doc; the team also says Euler started via a university hackathon win, which supports a technical, non-anon origin. Independent coverage of the v2 relaunch also ties Bentley directly to the post-hack rebuild and notes a large security effort around v2. On credibility, the major positive is that the team survived and rebuilt after the 2023 exploit; the major negative is that the protocol already suffered a major hack, so operational risk is not theoretical. Publicly available sources suggest a London headquarters / remote team footprint, but a real office, exact onshore/offshore corporate structure, and whether the business is primarily UK/onshore vs. offshore are not reliably verifiable from the gathered sources. Likewise, the team looks public rather than anonymous, but the precise legal entity and domicile should be treated as not verifiable as of 2026-08-29. In risk terms: credible technical founders, public identities, prior exploit history, and insufficiently verified corporate-jurisdiction detail; that is a meaningful but not clean institutional profile.

Euler V2’s reputation is mixed but improving, shaped by the major Euler v1 exploit in March 2023 and a heavily publicized “security-first” relaunch with extensive auditing and testing. Exploit history & incident handling

• In March 2023, Euler v1 suffered a ~$200m exploit via a smart contract vulnerability; this is a material adverse event for reputation.
• The attacker ultimately returned most of the stolen funds after negotiations; Euler then allowed users to redeem recovered funds, which mitigated user losses and reputational damage.
• Euler Labs publicly acknowledged that the vulnerable code had been audited but the issue was missed, which raised concerns about prior security processes but also about transparency. Security posture and audits for Euler V2
• For V2, Euler DAO reportedly allocated about $4m to security and completed 29–31 audits by ~12–16 security firms, including Certora, Trail of Bits, OpenZeppelin, ChainSecurity, Omniscia, OtterSec, yAudit, Zellic and others.
• Reports indicate two critical and five high‑severity issues were found in V2, all claimed to be fixed before launch.
• A large Cantina audit competition with $1.25m rewards and a Hats Finance “capture the flag” with about $3.5m at risk reportedly found no medium‑or‑higher severity issues in the live contracts.
• The team highlights Certora formal verification of a “Holy Grail” safety property for the V2 vault; this is positioned as a control that would have prevented the v1 bug. These are still *unverified marketing claims* until independently replicated. External risk / reputation assessments
• DeFi Sentinel (independent analytics) gives Euler an AA rating with an 85/100 safety score and characterizes risk as “very low,” citing multiple audits and structured governance.
• Another independent audit-style report (Envelop/Index) describes Euler as “one of the most heavily audited protocols” but flags centralization and tokenomics risk, noting a high share of tokens allocated to insiders/investors.
• An OG “trust score” on social media rates $EUL relatively mid-range (40.36 / 100), suggesting moderate rather than top‑tier perceived trust. Regulatory / legal / sanctions context
• The 2023 exploit investigation involved law enforcement and blockchain analytics firms; one AML-focused database links the incident to North Korea–associated Lazarus Group, emphasizing regulatory scrutiny around the hack.
• No public record in the retrieved data indicates Euler, its founders, or investors are subject to sanctions or formal regulatory enforcement actions as of this writing. Not verifiable as of 2026-08-29. Unresolved concerns for an institutional risk lens
• Historical exploit risk (v1) remains a reputational overhang despite high audit coverage in V2.
• Governance and token concentration risks are flagged by independent analysts, implying potential influence by founders/investors.
• The security narrative for V2 is intensive, but the absence of on-chain verification and full public disclosure of all audit reports means residual model and implementation risk remains. Not verifiable as of 2026-08-29.

Euler V2 is a modular, permissionless lending protocol built around isolated ERC‑4626 vaults, where each vault holds one asset and defines its own risk and interest‑rate parameters. Strategy & assets in/out

• Assets in: Users deposit ERC‑20 tokens into lending vaults and receive ERC‑4626 vault shares.
• Assets out: Borrowers draw other assets from connected vaults, subject to collateral relationships defined by curators/market creators.
• Markets are formed by explicit collateral links between vaults, chosen oracles, interest‑rate models, and caps. Yield source: organic vs subsidized
• Core yield is organic lending interest: borrowers pay variable rates to lenders, determined per‑vault IRMs (interest‑rate models) based on utilization.
• No systemic protocol‑wide subsidy is described; any incentive programs would be vault/chain‑specific and not verifiable as of 2026‑08‑29 from current data. Risk profile: market‑neutral vs directional; leverage/looping
• Lenders are directional on the deposited asset’s price plus borrower credit/liquidation risk.
• Borrowers can construct leveraged and looping positions by borrowing against collateral in other vaults via the Ethereum Vault Connector (EVC), enabling multi‑vault strategies and leverage similar to Aave/Compound loops.
• The system itself is not market‑neutral; it is generic credit infrastructure where users can build both hedged and directional strategies. Lock‑ups, withdrawals, mechanics
• Vaults are ERC‑4626 style and do not have protocol‑level time lock‑ups; deposits/withdrawals are on‑demand, constrained only by available liquidity and per‑vault caps.
• Liquidations are triggered when an account’s cross‑vault collateralization falls below thresholds; v2 supports custom and more gas‑efficient liquidations, including partial liquidations. Fees, gates, limits, revenue
• Protocol revenues come from spread/fees embedded in interest‑rate models or reserve factors at vault level (standard money‑market design); exact take‑rate per vault is parameterized and not summarized at system level in current public docs.
• Each vault may have supply/borrow caps and risk parameters (LTV, liquidation bonus), effectively gating maximum exposure per asset/market. Collateral model
• Any listed asset can become collateral via explicit collateral relationships between vaults; risk is isolated per vault rather than a single global pool. Chains and TVL
• Aggregators report Euler V2 active on multiple chains (≈10–16) with TVL around $0.5–1.0B across >100 pools, but exact chain‑by‑chain TVL and product breakdown are inconsistent across sites and not verifiable as of 2026‑08‑29.
• Historical APY data (e.g., DefiLlama: 125 pools, avg APY ~6–7%) shows vault‑ and chain‑specific volatility, with yields varying by utilization and asset; long‑term sustainability depends on organic borrowing demand, not fixed emissions. Contradiction callout (TVL)
• One analytics source shows ~$518M TVL from a DeFiLlama feed with 16 chains, while another independent analytics profile cites ~$982M TVL across 10 chains with 111 pools. These rely on different aggregation methods; without on‑chain queries, the true figure is Not verifiable as of 2026‑08‑29.

Euler V2’s publicly documented treasury is the Euler DAO Treasury at 0xC7C5aFDB61e08BE3e2FB09098412b5706EB5c550, controlled by the Euler Foundation and used for asset management and allocation; Euler also states the treasury’s asset set includes EUL, USDC, ETH, and sometimes airdropped tokens such as GRT. The docs also describe a November 2025 transition in which the old treasury multisig was recategorized as an operational multisig and treasury assets were transferred to the new treasury address, with the legacy address remaining relevant only as a historical/operational reference.

Euler V2 does have a native token: EUL (ERC-20), which is the protocol’s governance token and is also used for platform rewards and as the payment token in Fee Flow auctions; Euler’s docs state Fee Flow is currently inactive because protocol fees are set to 0%. The documented total supply is 27,182,818 EUL. I cannot verify the current circulating supply, market cap, FDV, top-holder concentration, insider wallets, mint/blacklist/fee-switch controls, or whether announced unlocks actually occurred on-chain in this run; those are Not verifiable as of 2026-08-29. The same applies to a chain-by-chain DEX liquidity breakdown and the main listings beyond the limited liquidity snapshots surfaced by aggregators. Known tokenomics from the gathered sources:

• Utility/governance: voting power in Euler DAO governance, rewards, and Fee Flow payment token.
• Unlocking/vesting: rEUL converts 1:1 into EUL over six months; 20% unlocks immediately and the remaining 80% vests linearly over six months, with forfeited unvested EUL sent to the configured remainder receiver.
• Allocations: one surfaced tokenomics source attributes allocations to Investors (39.5%), Community (26%), Insiders (20.6%), and Foundation (13.8%). Another surfaced vesting source gives a different breakdown, including Euler Labs Shareholders, Community Selected Markets, Employees and Advisors, Treasury, EulerDAO partners, Project Incubators, Binance HODLer Airdrops, and Retroactive Distribution. These figures conflict and are not reconciled here.
• Liquidity/listings: a surfaced aggregator shows very low EUL liquidity on Ethereum pools, including Uniswap V3 and Balancer V2, but I cannot validate broader venue coverage or multi-chain depth here. No source gathered here supports a claim that EUL has revenue share, buybacks, or burns as an active mechanism; Euler’s docs instead say Fee Flow is currently inactive with fees at 0%.

Euler V2 is a modular, vault-based lending protocol, so a Bitcoin move below $10,000 would not create one uniform outcome across the whole system; the impact would depend on each vault’s collateral type, oracle, and liquidation thresholds. The protocol’s design explicitly evaluates solvency at the account level across connected vaults, and positions that fail the configured collateral requirements can be liquidated. For BTC-linked collateral or debt, a crash below $10,000 would likely trigger rapid liquidations wherever borrowed positions are leveraged against BTC exposure, because Euler liquidations are onchain and automatic once account health is breached. Euler’s own postmortem on the Dec. 2024 Bitcoin flash crash shows that sharp BTC price dislocations can liquidate leveraged BTC positions when wrapped-BTC assets and oracle pricing diverge under stress. That indicates the main risk is not only the spot BTC move itself, but also basis dislocation, oracle lag, and wrapper-specific pricing differences. For non-BTC vaults, the effect is indirect and depends on whether BTC is used as collateral, whether that collateral is revalued through a shared oracle path, and whether the vault’s risk parameters are conservative enough to absorb volatility. Because Euler V2 isolates risk by vault, stress in one market is intended to remain contained rather than propagate system-wide. What is not verifiable as of 2026-08-29 from the provided sources is the protocol’s current chain-by-chain exposure to BTC, the share of TVL in BTC-related markets, and whether any specific Euler V2 vault on Arbitrum, Avalanche, BSC, Base, Ethereum, Hyperliquid L1, Linea, Monad, Plasma, or Unichain has BTC exposure large enough to threaten broader losses. The available sources do not provide the necessary onchain data, and Dune-based verification is unavailable in this run. The practical risk takeaway is: BTC below $10,000 would likely cause severe, localized liquidation pressure in BTC-related Euler markets, but protocol-wide contagion is structurally limited by vault isolation.

For Euler V2, a stress case where the largest collateral depegs by 20% is not verifiable as of 2026-08-29 with the available sources, because no current, protocol-wide, chain-by-chain collateral composition or position data is available here. What can be said from the retrieved material is that Euler’s liquidation framework is designed to absorb underwater positions through liquidations, with a reserve fund as a last resort if collateral sale does not cover debt; the protocol also documents liquidation LTV ramps and account-level solvency checks. The only direct 20% depeg-related evidence in the sources is historical: Euler’s incident/post-mortem and Coinbase’s investigation note that Euler’s liquidation bonus was capped at 20% in the 2023 exploit context, which is not the same as stress-testing today’s collateral base. Because the question depends on the largest collateral asset(s) on Euler V2 across Arbitrum, Avalanche, BSC, Base, Ethereum, Hyperliquid L1, Linea, Monad, Plasma, and Unichain, the key inputs needed to quantify losses would be each market’s exposure, LTV/liquidation LTV, oracle design, and whether the asset is used as collateral in active vaults. Those inputs are not verifiable from the provided sources alone. In short: a 20% depeg of the biggest collateral would likely trigger liquidations and could create bad debt if the liquidation discount, liquidity, or reserve coverage are insufficient, but the size of any loss is Not verifiable as of 2026-08-29.

Euler v2 is vault-based and modular, so a “top counterparty insolvent” scenario plays out *per vault* and then propagates via the Ethereum Vault Connector (EVC) where collateral is shared across vaults. Because Dune/on‑chain is unavailable: all chain allocations, TVL shares, and exact sizes of counterparties are “Not verifiable as of 2026‑08‑29”. ### 1. Base mechanics: how insolvency appears

• Each EVK vault is an ERC‑4626-style lending vault with its own asset, interest model, oracle, and risk params.
• EVC lets positions in one vault serve as cross‑vault collateral, enabling multi‑vault strategies and cross‑vault liquidations.
• A “top counterparty” here is a large borrower (or leveraged strategy) whose liabilities sit in one or more vaults. Insolvency path (per chain): 1. Shock: price move or oracle/manipulation event makes the borrower under‑collateralised in one or more vaults. 2. Liquidation attempt: liquidators use the EVC to seize collateral from connected vaults and repay debt where profitable. 3. If collateral is inadequate or liquidations fail (oracle pause, liquidity gap, gas spikes), the vault accrues bad debt: liabilities > assets. ### 2. Who absorbs the loss? Per affected vault on a given chain:
• Lenders/depositors in that vault bear first loss: vault share price effectively reflects less underlying than expected (economic haircut to LPs).
• If EVC‑connected vaults were used as collateral, liquidations can propagate and drain value from other vaults that held the insolvent borrower’s collateral, spreading losses across those depositor sets.
• There is no protocol‑level seniority waterfall across independent EVK vaults: loss allocation is determined by how the vault was designed (e.g., isolated vs cross‑collateral, curator rules) and by EVC liquidation logic. ### 3. Compensation / backstops
• Public sources describe heavy auditing and risk work, but do not document an on‑chain insurance fund, backstop module, or automatic recapitalisation for Euler v2.
• Any compensation in a severe insolvency is therefore off‑chain / social (governance, foundation decisions, or OTC recap), not a guaranteed smart‑contract feature. This is not verifiable on‑chain as of 2026‑08‑29. ### 4. Impact path through contracts (per chain) For Arbitrum, Avalanche, BSC, Base, Ethereum, Hyperliquid L1, Linea, Monad, Plasma, Unichain:
• If Euler v2 is deployed on that chain and the insolvent counterparty is active there (TVL share per chain is Not verifiable as of 2026‑08‑29), the path is:
• Borrower health factor breach in local vault → EVC‑mediated liquidations → collateral sales / transfers into liquidator vaults → residual bad debt remains in one or more vaults → economic loss to those vaults’ depositors.
• Contagion is bounded by EVC connectivity: vaults not connected as collateral to the failing positions should not directly absorb losses, but may face indirect effects (liquidity flight, governance responses, parameter changes). No automatic protocol‑level make‑whole is documented; large LPs must treat each EVK vault + its EVC exposure graph as the unit of loss in a top‑counterparty default scenario.

For Euler V2, a stress scenario of committed fraud by the DAO or owners is not verifiable as of 2026-08-29. I did not find credible evidence in the supplied sources that Euler V2’s DAO or owners committed fraud; the strongest relevant material instead points to a past exploit in Euler V1 caused by a protocol vulnerability, not by proven DAO/owner fraud. What can be supported is that Euler’s design includes centralization and governance risk: one audit-style report notes concentrated token ownership and warns that complex architecture can create risks from malicious controllers or operators, which is relevant to governance abuse scenarios but is not itself evidence of fraud. Euler’s own retrospective also shows the DAO can make major strategic decisions, including recommending sunset of directly managed markets and vaults, which indicates governance authority but does not demonstrate misconduct. The important distinction is that the documented 2023 incident involved an exploitation of protocol logic in Euler V1, with public post-mortems describing a missing health check around donateToReserves and resulting liquidation abuse; those sources describe a hack and recovery process, not a fraud finding against the DAO or owners. If you want, I can next assess whether governance capture, insider allocation, or admin-key abuse creates a fraud-like risk profile for Euler V2 across its supported chains, but that would remain a risk assessment rather than a verified fraud event.

Euler V2’s primary yield source is borrower interest, so a negative 30d yield stress scenario means the protocol’s deposit-side return over the last 30 days is below zero only if net fees/rebates/liquidity effects outweigh accrued borrowing income; the provided sources do not give a 30-day yield series, so the actual negative-30d condition is Not verifiable as of 2026-08-29. Euler V2’s design states that suppliers earn yield from borrower interest, while borrow interest accrues to the liability vault and then flows to depositors after configured fees, so sustained negative depositor yield would indicate an unusual stress condition rather than the normal operating model. The protocol also emphasizes solvency checks across EVC-registered vaults, liquidation when collateral falls short, and bad-debt/socialization mechanics, which are relevant because severe undercollateralization can impair depositor returns. For the requested chain set (Arbitrum, Avalanche, BSC, Base, Ethereum, Hyperliquid L1, Linea, Monad, Plasma, Unichain), chain-by-chain exposure and contribution to the 30d yield cannot be verified from the supplied sources, so each chain’s share is Not verifiable as of 2026-08-29.

Euler v2 appears to be governed primarily by the Euler DAO, with substantial ongoing control by the founding company and core contributors. On-chain verification is not possible in this run; any such facts are "Not verifiable as of 2026-08-30". 1. Governance structure & who controls what

• Protocol development & contracts: Euler v1 was built and maintained by Euler Labs Ltd (UK-based); Euler v2 continues to be led by the same core team and contributors. Control over upgrades is reportedly via governance contracts and multisigs, but exact v2 addresses and roles are Not verifiable as of 2026-08-30.
• Frontend & branding: The main interface and branding are operated by the Euler core team / Euler Labs, not directly by the DAO. This implies company-level control over frontend availability and routing.
• Treasury/funds: Euler DAO treasury assets and protocol fees are governed by DAO votes, with execution via multisig signers controlled by elected or core contributors. Specific treasury addresses, balances, and spend policies are Not verifiable as of 2026-08-30. 2. DAO reality vs symbolism
• Historical governance for Euler used token-based voting, snapshot proposals, and on-chain execution for certain parameters, but the core team retained substantial influence through proposal drafting, multisigs, and technical control of deployments.
• There is no evidence that Euler v2 is fully autonomous; governance appears mixed: DAO-guided but company- and team-steered, with meaningful discretion by core contributors over dev roadmap, security response, and frontend operations. 3. Governance process (proposals & voting)
• Euler governance follows a familiar DeFi pattern: community discussion → off-chain signaling (e.g. Snapshot) → on-chain governance proposal → execution by contracts/multisigs. Detailed v2 parameters, quorums, and thresholds are Not verifiable as of 2026-08-30.
• Voting power is based on EUL governance tokens; distribution, concentration, and top holders (normally checked via Dune) are Not verifiable as of 2026-08-30. 4. Timelock & multisigs
• Prior Euler governance used timelock mechanisms for upgrades and risk parameter changes, plus multisigs with multiple signers including core team members and sometimes community representatives. Exact v2 timelock durations, signers, thresholds, and independence are Not verifiable as of 2026-08-30. 5. Legal entity, jurisdiction, ToS
• Euler Labs Ltd has been referenced as a UK private limited company behind Euler protocol development. Full company registration number, current directors, and exact ToS governing Euler v2 are Not verifiable as of 2026-08-30. Key risk takeaway: governance for Euler v2 is not credibly fully decentralized; critical powers over dev, upgrades, and frontend remain concentrated with the founding company and core contributors, with a token-based DAO providing direction but not exclusive control.

Euler V2 itself is a lending protocol, so the answer depends on which stablecoin market you mean. The web results show at least one clear Euler-wide stablecoin stress event: a March 11, 2023 monitoring thread on the Euler forum about the USDC depeg, which also noted DAI was affected and that the team was monitoring protocol-wide risk metrics. However, with the current evidence, it is not verifiable as of 2026-08-29 whether the specific stablecoin used in your chosen Euler V2 market depegged, how many times it happened, the last occurrence, or the exact depeg percentage. The only concrete depeg-related incident surfaced here is the USDC event referenced above; a separate forum post also mentions FDUSD sustaining a depeg event in April 2025, but that appears to be about Euler market offboarding rather than a core Euler V2-wide stablecoin history. If you want a precise answer, the market-specific stablecoin symbol or vault address is needed.

Euler V2’s top risks are smart-contract/architectural complexity, oracle and collateral-link risk, cross-vault contagion, governance/configuration risk, and historical/regulatory reputational risk. The protocol itself says V2 is experimental, can suffer smart-contract failures, cyberattacks, unknown vulnerabilities, and unforeseen interactions that may cause partial or total asset loss. Hindenrank highlights the permissionless design, noting that anyone can create a lending market and link it to others, which expands attack surface and can make governance takeover cheaper than in more mature systems. Electisec identified a critical issue where deloop() lacked a controller check, enabling bad debt creation between mutually linked vaults, plus a high-severity reentrancy issue in EulerSavingsRate. Euler’s own docs also emphasize that risk is driven by each vault’s live configuration, including collateral links, oracle routes, caps, LTVs, and governance settings, so misconfiguration can directly affect solvency. Finally, Euler V2 still carries reputational overhang from the 2023 Euler v1 exploit, and third-party reviews flag token concentration and governance-capture concerns as additional non-technical risk factors.

Euler V2’s main strengths are its modular, permissionless vault architecture, risk isolation, cross-vault composability, customizable market parameters, and strong security design. The protocol is built around independent vaults created with the Euler Vault Kit (EVK), so markets can be tailored per asset and risk profile instead of forcing users into one shared pool.