In the discussion, ChaosLabs introduced a concept called Isolation Mode for the Aave protocol, aimed at mitigating the risks associated with catastrophic asset devaluation events. The proposed tool works by setting a debt ceiling, which is the maximum amount of debt that can be borrowed against an isolated asset. The debt ceiling is determined through a stress-testing framework, which considers the on-chain liquidity of tokens and simulates the behavior of profit-maximizing liquidators. The potential losses from an isolated asset are limited to a certain percentage of the Safety Module (SM), with the percentage being more conservative or aggressive based on the perceived riskiness of the isolated asset¹.

ChaosLabs further provided an estimation of the losses to the Aave protocol as a function of the debt ceiling. This was achieved by simulating a profit-maximizing liquidator transacting with the LUNA/ETH Uni v3 pool during the LUNA crash, and then scaling this loss curve by a liquidity ratio¹. The guidelines proposed for setting the debt ceiling are: 1% of the SM for risky assets and 5% of the SM for less risky assets. Using these guidelines, the liquidity of each token, and the LUNA price/liquidity trajectory during its crash, ChaosLabs provided the debt ceiling results for MKR, UNI, and SNX tokens¹.

In conclusion, ChaosLabs emphasized that predicting a catastrophic asset devaluation is a complex problem, akin to predicting black swan events. Therefore, they recommend setting conservative debt ceilings by estimating losses on isolated assets using the LUNA collapse and capping the potential losses to Aave at 1% or 5% of the safety module¹. This approach aims to protect the Aave protocol from significant losses in the event of a severe asset devaluation.