Summary
Ketju does not approve Euler V2 as a protocol: each vault sets its own collateral, oracle, governor and upgrade rules, so the name Euler tells a client little about the risk, and Ketju will review only a named vault. The original Euler lost about $197M in March 2023 and recovered 100% of the stolen assets, worth about $240M when returned. Euler V2 launched in September 2024 after a rebuild from scratch. Those facts count, but they do not support approving the whole protocol. Euler did not report $137M of protocol bad debt from Stream Finance. Its 2026 retrospective says DAO-managed markets on Ethereum had no direct exposure to Stream. The incentivized Plasma deployment lent heavily to Stream and was effectively wiped out. That was a serious failure of allocation and governance, though no second exploit of the core contracts. Each EVK vault chooses its collateral, LTVs, interest-rate model, oracle, hooks, governor and upgrade setting. EulerEarn can allocate across as many as 30 ERC-4626 strategies, including outside ones. Euler's docs say a “known” label on a vault is no endorsement, and curators can change parameters later. A named vault needs its own memo, built address by address; brand, total TVL and the number of audits cannot stand in for it.
Review triggers
- No approval for the whole protocol: Ketju reviews only a named vault
- Any candidate vault holds or accepts an asset outside the approved list
- Any realized bad debt, oracle failure or loss in the candidate vault or an underlying strategy
- The governor, curator, strategy, oracle, implementation or upgrade setting changes outside the documented window for holders to withdraw
- A withdrawal of the proposed size fails the written stress-liquidity limit
Facts
- Verdict
- Not approved
- Type
- Dollar lending
- Chains examined
- Ethereum
- Instruments
- USDC, USDT, WETH
- Reviewed
- Last confirmed