LlamaRisk Proposes AAVE V3 Core Freeze Covering $538.6 Million in Collateral

LlamaRisk's Tuesday Proposal

Aave could freeze three more collateral tokens on its largest Ethereum lending market under a proposal the risk service LlamaRisk filed on Tuesday, Nov. 10, and any Aave price reaction rides on a vote that, as of Tuesday, had not been scheduled. LlamaRisk, an adviser funded in part by the Aave DAO, asks the protocol to freeze StakeWise's osETH, a staked ether token; Lombard's LBTC, a yield-bearing bitcoin token; and Rocket Pool's rETH, a liquid staking token, along with six small reserves. The stated aim is to keep Aave V3 Core, the protocol's main Ethereum market, reserved for established and liquid collateral, and to move riskier assets toward Aave V4. The freeze would apply to the Core market specifically, not to every deployment of these tokens across Aave. Five assets appear in the proposal, but two are already frozen: rsETH, Kelp DAO's restaked ether token, since April, and tBTC, Threshold Network's bitcoin token, since Oct. 31, after Threshold paused minting and redemptions and cut the bridge asset's convertibility. Measured on LlamaRisk's Nov. 9 figures, the three largest new freezes cover $538.6 million of collateral backing $364.1 million of debt. Everything in scope adds up to $1.33 billion of collateral standing against $1.03 billion of debt, and the DAO earns roughly $3.7 million a year in revenue from that borrowing. The measure is an Aave Request for Final Comments, a stage that precedes any binding vote, and discussion is still open on the Aave governance forum.

A freeze under Aave's rules is a stop on new risk, not a seizure of funds. It blocks fresh deposits and new borrowing against a token and sets its loan-to-value ratio, the share of a deposit's value a user can borrow, to zero. Suppliers in the affected liquidity pools can still withdraw. LlamaRisk CEO Adam Sadowski, discussing the protocol's Oct. 29 emergency freeze in early November, called the tool “a one-way valve”: users can always exit, while no new exposure can come in. The filing says existing positions stay “untouched and unaffected.” Most of the new exposure sits in osETH, where $279.4 million of collateral backs $256.6 million of wrapped ether (WETH) loans, a loop running at roughly 14 times leverage, closer to repeated margin trading than to plain deposit-and-borrow use. The plan also raises the reserve factor, the cut of borrowers' interest routed to the DAO rather than to lenders, to 99% on seven borrowable frozen reserves that still carry more than $20,000 of debt, leaving borrow rates unchanged. If exposure does not shrink on its own, the proposal authorizes stepwise cuts to liquidation thresholds, the collateral level at which a loan can be liquidated, and, only as a final step, a higher WETH borrow rate aimed at the rsETH and osETH loops. LlamaRisk pledges not to make “abrupt parameter changes” that would liquidate significant positions. The approach echoes the protocol's recent crisis handling: after the $6 billion Binance hack ran through Aave, lending markets were reopened with nine assets at zero LTV, and LlamaRisk later set out a four-stage plan to reopen markets.

No Snapshot Vote Yet

Nothing in this file is decided. As of Nov. 11, no ballot on the ARFC had been posted to the DAO's space on Snapshot, the offchain voting platform, and a poster signing as StakeWise's founder asked that day for the proposed osETH freeze to move to a separate vote, adding a procedural fork before any tally. Holders of the governance token face a trade the figures frame cleanly: about $3.7 million a year of DAO revenue against $1.33 billion of collateral in assets the adviser treats as less established. Aave Labs' Avon acquisition, which brings fixed-rate lending into the protocol's product set, shows the same repositioning running in parallel. The matter closes when a Snapshot ballot posts and passes, and the record, as of Tuesday, does not yet fix when that happens.