Kelp DAO Hack Explained: How a $292M Bridge Exploit Happened

Key Takeaways

• The Kelp DAO hack drained roughly $292 million (about 116,500 rsETH) on April 18-19, 2026, making it the largest DeFi exploit of the year and the second-largest bridge hack in crypto history after Ronin.
• Attackers didn’t break Kelp’s smart contracts or steal a private key in the traditional sense — they forged a cross-chain message that Kelp’s LayerZero-powered bridging adapter accepted as genuine.
• The root cause was a “1-of-1” verifier configuration: Kelp’s deployment required only one party to attest that a message was legitimate, so compromising that single off-chain verifier was enough to drain the bridge.
• LayerZero preliminarily attributed the attack to North Korea’s Lazarus Group, whose off-chain infrastructure compromise let them impersonate a trusted attestor.
• Kelp’s team detected the anomaly and paused contracts within roughly 46 minutes, blocking a follow-up attempt to drain another 40,000 rsETH, and Arbitrum’s Security Council later froze about $71 million linked to the theft.

Article Summary: The Kelp DAO hack was a $292 million cross-chain bridge exploit in April 2026 caused by a forged LayerZero message accepted by a bridge that relied on a single verifier, not a smart contract bug or a stolen validator key, making it the clearest example yet of how message-forgery attacks differ from the bridge vulnerabilities investors are used to watching for.

What Is Kelp DAO?

Kelp DAO is a liquid restaking protocol built on Ethereum. Users deposit staked ETH derivatives and receive rsETH in return — a liquid token that represents their restaked position while remaining tradable and usable as collateral across DeFi. To make rsETH useful outside Ethereum, Kelp deployed bridging infrastructure using LayerZero, a widely adopted cross-chain messaging protocol, so rsETH could move to and be used natively on more than 20 other blockchains.

That reach is exactly what made the April 2026 exploit so damaging. The bridge being drained held the reserves backing wrapped rsETH deployed across those 20-plus chains, which meant every protocol accepting rsETH as collateral was suddenly exposed the moment the reserves were compromised.

Timeline: How the Kelp DAO Hack Unfolded

On April 18-19, 2026, an attacker drained approximately 116,500 rsETH, worth roughly $292 million at the time, from Kelp DAO’s bridging adapter.¹ The withdrawal happened through a forged cross-chain message rather than any interaction with rsETH’s token contract directly, which is why on-chain monitoring tools built to flag suspicious contract calls didn’t immediately catch it as an exploit in progress.

Kelp’s team detected the anomaly and paused the relevant contracts, and their blacklisting of the attacker’s addresses blocked a follow-up attempt to drain a further 40,000 rsETH.¹ Arbitrum’s Security Council later froze roughly 30,766 ETH, worth about $71 million, that investigators linked to the theft.¹ Recovery efforts continued for weeks afterward. Downstream, lending markets moved fast: Aave froze its rsETH markets on V3 and V4 within hours, and SparkLend and Fluid followed with their own freezes as depositors rushed to pull funds from anything with rsETH exposure.²

How the Attack Actually Worked

To understand the Kelp DAO hack, it helps to understand what a cross-chain bridge is actually doing under the hood. A blockchain is a closed system — Ethereum can’t natively read what happened on Arbitrum or any other chain. Bridges solve this by having off-chain participants watch the source chain and then attest, or “sign off,” that an event genuinely occurred, so the destination chain knows it’s safe to release or mint funds.

Kelp’s deployment used LayerZero to handle that attestation process. LayerZero’s architecture allows a project to choose how many independent verifiers, known as Decentralized Verifier Networks or DVNs, must agree before a message is treated as genuine. Kelp’s relevant configuration used just one: a “1-of-1” setup where a single verifier’s signature was sufficient.³

The attacker didn’t need to break LayerZero’s cryptography or find a bug in Kelp’s smart contracts. They compromised the off-chain infrastructure behind that one verifier and used it to forge a message instructing the bridge to release funds.³ From the bridge’s perspective, the message looked completely legitimate — it carried a valid signature from the party it was configured to trust. The smart contract did exactly what it was written to do. The failure lived entirely in the trust assumption underneath the code, not in the code itself.

This is the core distinction that makes the Kelp DAO hack different from the bridge exploits investors have historically been trained to watch for.

Kelp DAO vs. Other Bridge Hack Patterns

Ronin was a story about who held the keys. Wormhole was a story about a coding gap. Kelp DAO is a story about a configuration decision — how many independent parties a project chose to require before trusting a cross-chain instruction. That distinction matters because it means an audit clean bill of health tells an investor almost nothing about this specific risk. A bridge’s smart contracts can be immaculately audited and formally verified while its message-verification setup still has exactly one point of failure.

Why This Matters Beyond Kelp DAO

The Kelp DAO hack wasn’t an isolated misconfiguration. It was the largest example of a pattern that showed up repeatedly through 2026. In February, the cross-chain protocol CrossCurve lost nearly $3 million when attackers spoofed cross-chain messages through an Axelar-linked receiver contract with access-control gaps. In May, the Verus-Ethereum bridge lost roughly $11.6 million to a bridge that verified a proof existed but never checked whether the claimed amount matched it. In June, the Syscoin bridge lost about $10 million after an attacker submitted a fake burn proof that exploited a parsing flaw in the relay’s validation code. Security researchers tracked roughly $328.6 million in total cross-chain bridge losses across at least eight major 2026 incidents, and message-layer failures like Kelp’s now sit alongside smart contract bugs and key theft as one of the field’s dominant attack patterns. See the full, regularly updated list in our 2026 Bridge Hack Tracker.

For a fuller breakdown of how message forgery compares to the other major bridge vulnerabilities, including validator compromise, oracle manipulation, and centralization risk, see our complete guide to bridge risk in 2026.

What Investors and DeFi Users Should Check Before Trusting a Bridge

The Kelp DAO hack offers a concrete, practical checklist for anyone holding assets behind a cross-chain bridge, whether that’s a liquid restaking token, a wrapped asset, or bridged mining rewards.

Ask how many verifiers are required, not just whether the bridge was audited

Reputable projects using message-passing protocols like LayerZero or Axelar should be able to tell you exactly how many independent Decentralized Verifier Networks or equivalent parties must agree before a message is accepted. If that number is one, or if the project can’t or won’t answer the question, treat that as a serious red flag regardless of how polished the rest of the protocol looks.

Understand your indirect exposure

You don’t have to hold rsETH directly to be exposed to a hack like this. Lending markets that accept a bridged asset as collateral inherit that asset’s bridge risk the moment they list it. Aave, SparkLend, and Fluid all had to freeze rsETH markets within hours of the Kelp DAO exploit, and protocols were still quantifying bad debt from borrowers whose collateral had effectively become an unbacked claim. Before depositing into any lending market, check what bridged assets it accepts as collateral and how those assets get to the chain in the first place.

Watch how fast a protocol actually responds

Kelp’s roughly 46-minute response time, which stopped a second drain attempt, is genuinely fast by industry standards and worth noting as a point in the protocol’s favor going forward. But it’s also a reminder that detection speed is a mitigation, not a prevention. The first drain still happened. A fast pause limits follow-up damage; it doesn’t undo the initial loss.

Kelp DAO Hack FAQs

How much was stolen in the Kelp DAO hack?

The Kelp DAO hack resulted in the theft of roughly 116,500 rsETH, worth approximately $292 million at the time of the attack in April 2026, making it the largest DeFi exploit of the year.

Who was behind the Kelp DAO hack?

LayerZero preliminarily attributed the Kelp DAO hack to North Korea’s Lazarus Group, based on the attacker’s method of compromising the off-chain infrastructure behind Kelp’s single message verifier.

Was any of the stolen money recovered?

Arbitrum’s Security Council froze approximately $71 million linked to the theft, and Kelp’s own response blocked a follow-up attempt to steal an additional 40,000 rsETH, but the majority of the initial $292 million was not recovered.

Could the Kelp DAO hack have been prevented?

Requiring multiple independent verifiers instead of a single one would very likely have prevented the Kelp DAO hack, since the attacker would have needed to compromise more than one party’s off-chain infrastructure simultaneously rather than just one.

Kelp DAO Hack Citations

• Mintlayer, “Anatomy of 2026’s Bridge Exploits” – mintlayer.com
• Phemex, “Every Major DeFi Hack in 2026 So Far | Bridge Exploits Dominate” – phemex.com
• 1inch Blog, “The Biggest Bridge Hacks in 2026” – 1inch.com