LlamaRisk supports the proposal of @ChaosLabs for the dynamic calibration of CAPO parameters. The automation of maxYearlyRatioGrowthPercent and snapshotRatio parameters reduces the operational burden on service providers who would otherwise need to perform frequent manual updates. We acknowledge that the proposed methodology is designed to be a safe enhancement, tracking and updating bounds without fundamentally altering the core, battle-tested CAPO components.
The proposal correctly identifies the risks of stale parameters. A widening gap between the real exchange rate and the CAPO upper bound does not create the attack vector in itself, but it significantly exacerbates the potential damage from price inflation or donation attacks. We concur that maintaining tight bounds is crucial for protocol security. At the same time, there is a trade-off, as overly restrictive bounds could lead to a negative user experience. Specifically, if the maxYearlyRatioGrowthPercent is set too low, the oracle will cap the asset’s price below its true, organically grown exchange rate. This would cause a user’s collateral to be undervalued by the protocol, unfairly limiting their borrowing power.
The document outlines a decision rule for upward adjustments under rate escalation, which preemptively triggers a parameter update if the observed short-term yield velocity suggests the current cap could be breached before the next scheduled update window. This rational approach ensures the system can react to legitimate asset yield growth. While the proposal also mentions a trigger for downward adjustments based on longer-term historical data without specifying it, more transparency regarding the thresholds of this mechanism would also help to reason about the safety of the CAPO adjustments methodology.
Finally, the proposal introduces a conservative constraint on the maxYearlyRatioGrowthPercent, limiting updates to a maximum 10% relative change with a 3-day timelock. While this rate-limiting is a prudent safety measure against sudden, unwarranted spikes, it may prove insufficient during sustained, aggressive changes in an asset’s yield profile. When a legitimate yield source rapidly increases (e.g., sUSDe yield increases from 10% to 30% APY), the 3-day delay and 10% relative increase cap might cause the CAPO bound to lag behind the true organic growth. This parameter may require observation and potential tuning by governance as the system matures and encounters more diverse market conditions.
Disclaimer
This review was independently prepared by LlamaRisk, a community-led decentralized organization funded in part by the Aave DAO. LlamaRisk is not directly affiliated with the protocol(s) reviewed in this assessment and did not receive any compensation from the protocol(s) or their affiliated entities for this work.
The information provided should not be construed as legal, financial, tax, or professional advice.