LlamaRisk Supports Aave Plan to Deprecate 50 Reserves and Six Low-Adoption Markets

Deprecation plan targets long-tail exposure

According to the update, the low-adoption asset deprecation proposal is designed to reduce Aave’s risk surface rather than respond to a single incident. The package spans long-tail assets on Aave V3 and also includes 21 matured Pendle PTs, which LlamaRisk grouped into the broader clean-up effort.

The firm said the work was coordinated with a separate oracle deprecation effort for selected long-tail assets across Aave V2 and V3. In that track, LlamaRisk proposed moving affected assets toward fixed-price adapters and other controlled parameters as part of asset-specific migration plans, citing the operational risk tied to maintaining certain Chainlink price feeds.

For the Aave V3 assets and markets being deprecated, LlamaRisk said the implementation approach combines reserve freezes, reductions of supply and borrow caps to 1, and fixed-price oracles where relevant. The aim is to make exits orderly while limiting new activity in positions that no longer justify ongoing risk overhead.

Six markets marked for wind-down

Beyond individual reserves, the update identifies six low-adoption Aave V3 markets for wind-down: Sonic, Scroll, zkSync, Metis, Soneium and Aptos. LlamaRisk placed the combined footprint of those markets at $12.8 million in supplied assets and $4.1 million in debt.

The rationale presented in the monthly summary is straightforward: low usage can leave markets carrying infrastructure, oracle and monitoring complexity that outweighs their adoption. By concentrating liquidity and governance attention on better-used venues, the protocol can reduce maintenance burden while keeping the transition controlled through parameter changes rather than abrupt shutdowns.

Broader July governance work

The monthly report also lists several onboarding and product recommendations. LlamaRisk said it supported onboarding syrupUSDG on Aave V4’s Global Dollar Hub, contingent on sufficient decentralized exchange liquidity bootstrapping. It also updated its assessment for USDai and sUSDai on Aave V3 Arbitrum, saying support remained in place after the transition to PYUSD backing and KYC-restricted minting and redemption, while noting weaker secondary market liquidity for USDai.

Additional positive recommendations covered syrupUSDC on Aave V3 Core and PT-AUSD-8OCT2026 on the Aave V3 Monad instance. For syrupUSDC, LlamaRisk cited improved redemption times, six audits, a 24-hour timelock on contract upgrades and stable DEX liquidity. For the Pendle PT listing on Monad, it said the long maturity horizon supported onboarding despite limited automated market maker liquidity of $5.7 million, and it recommended an updated E-Mode borrowable set of USDT0, USDC, GHO and USDe.

The group also backed the launch of cross-chain sGHO, describing a design that would use Chainlink CCIP bridging while keeping yield generation and accounting unified through a single ERC-4626 vault. In a separate governance track, it supported the next stage of SVR expansion to Avalanche, Polygon and BNB Chain using Chainlink’s Atlas auction infrastructure with a 10-second fallback delay.

Risk operations and what comes next

Outside formal deprecation work, LlamaRisk said its risk stewards handled multiple supply cap, borrow cap and PT parameter changes across Aave V3 and V4 during July. It also reported additional rounds of V4 add-and-draw cap increases across Ethereum hubs including Core, Prime, Plus and Global Dollar, as well as Avalanche Core, lifting the total supply cap ceiling to about $802 million.

On the tooling side, the team added Monad support to the Aave Risk Dashboard, launched an Aave V4 Explorer for tracking hub-and-spoke liquidity flows and expanded the SVR Dashboard to cover the Aave V3 Monad instance. Looking ahead, the update says LlamaRisk plans to keep monitoring Aave V4 adoption, continue liquidation parameter tuning, extend SVR analytics to Avalanche, Polygon and BNB Chain, pursue RWA due diligence and develop a standard methodology for bridge rate limits with bridge providers and asset issuers.

For now, the clearest confirmed next step is continued governance follow-through on the deprecation and oracle proposals already outlined in July, with day-to-day risk parameter management across Aave V3 and V4 remaining part of the team’s ongoing mandate.