Curve Finance Bad Debt Pool Targets $700K LlamaLend Hole

Curve Finance founder Michael Egorov has proposed a market-based fix for roughly $700,000 of bad debt sitting in the protocol’s LlamaLend lending product, packaging distressed lender positions into a tradable token routed through a dedicated Curve StableSwap pool. Egorov announced the plan on April 26, 2026, saying the pool is permissionless and the Curve DAO is “invited but not required” to vote.

The structure would let trapped lenders sell their claims at a discount rather than wait on a Curve treasury rescue. It is also a working counterpoint to the coordinated industry effort assembling for Aave, where losses tied to the Kelp DAO bridge exploit reach as much as $230 million and have already drawn pledged contributions from Mantle, Lido, EtherFi, Aave DAO, and Aave founder Stani Kulechov.

Inside the LlamaLend hole that opened in October 2025

The shortfall lives in LlamaLend’s CRV-long market, which lets users borrow Curve’s crvUSD stablecoin against CRV, the protocol’s governance token. During the October 10, 2025 market crash, CRV fell faster than LlamaLend’s automated liquidation system could process. Lender deposits were left around 70% backed, and the affected lenders cannot fully withdraw. It is not the first such episode tied to Egorov’s own positions: a separate June 2024 cascade created roughly $10 million of temporary bad debt, which was paid off within days.

How Egorov’s option-style payoff would work

Egorov has already deployed a Curve StableSwap pool centered around 71% solvency, with a 1% swap fee, where the impaired vault tokens can change hands at a discount. Buyers take a long bet on CRV. According to Egorov’s math reported by CoinDesk, the bad debt begins to clear when CRV crosses about $0.96, and full recovery happens near $1.24. CRV traded near $0.23 on April 27, 2026.

“I propose a free-market based method of recovery with option-like payoff, working as an investment for everyone who wants to participate in the effort,” Egorov wrote in the governance post. He framed the asymmetry plainly: if CRV rises, positions deliquidate. If CRV falls, the collateral has already been converted to crvUSD, so the vault is no less backed.

Liquidity providers in the new pool would earn swap fees and any CRV incentives the DAO chooses to allocate. Admin fees would partly accrue in the distressed vault token itself, slowly moving some of the bad debt onto Curve’s books through trading.

What Aave’s $230 million Kelp DAO problem reveals

The Curve proposal landed during a parallel debate at Aave. Earlier in April 2026, an attacker exploited Kelp DAO’s LayerZero bridge and released 116,500 unbacked rsETH worth about $292 million, then borrowed real WETH against it on Aave. The response is a recovery effort called DeFi United, which has raised about $160 million of the roughly $200 million needed.

Egorov’s pitch points the other way. “I proposed a recovery mechanism of bad debt, which is not a donation but an investment vehicle for everyone who participates,” he wrote. Reception has been mixed. One forum commenter argued “the reality is that no one will buy the affected positions, because they generate no yield.” Others said the discounted vault token resembles a long-dated option on CRV.

What happens next at gov.curve.finance

The Curve DAO has not yet voted on whether to allocate CRV incentives to the pool. Because the StableSwap pool is already live, the plan does not require DAO action to function. A formal vote on incentives would set the terms for liquidity providers and decide whether admin fees route impaired tokens back to the DAO treasury. If the pilot clears the LlamaLend hole, Egorov said, the structure could be applied “in similar difficult situations” at Curve or other protocols. Aave’s Kelp DAO recovery is the obvious test case.