Skip to content

Conversation

@clmikestone
Copy link
Contributor

Rationale

Chaos Labs has deployed Risk Oracle infrastructure that secures Aave from various economic risks. Below is a breakdown of the key oracles Chaos operates for Aave today, their parameter scope, and the systemic risks they mitigate.

Chaos Risk Oracle Details


Risk Oracle:
Dynamic Slope 2

Function:
Dynamically adjusts the borrowing cost of yield-bearing assets in the event of sudden liquidity shocks

Parameters Set:
slope2

Assets Secured:
USDC, USDT, USDe, and wETH on Ethereum Core and Linea

Full list here

Value Secured as of Oct 30, 2025 (% of Aave TVL)*:
*(Value secured reflects distinct markets on Aave with no overlap)
≈ $22B (34%)

How TVS is Secured (Failure Scenario / Loss Quantification):
Liquidity shocks without dynamic interest rate adjustment can result in significant unwinding and bad debt accumulation.
Misconfigured slope2 for markets ≥UOptimal can push borrowing costs to the extreme, forcing mass unwinding of positions

Documentation/Proof:
Approved Governance Proposal

Contracts:
0x7ABB46C690C52E919687D19ebF89C81A6136C1F2


Risk Oracle:
CAPO Risk Oracle

Function:
Governs price cap on yield-bearing assets relative to base asset

Parameters Set:
maxYearlyRatioGrowthPercent, snapshotRatio

Assets Secured:
LSTs, LRTs across all Aave Instances

Value Secured as of Oct 30, 2025 (% of Aave TVL)*:
≈ $20B (31%)

How TVS is Secured (Failure Scenario / Loss Quantification):
Governs asset prices; misconfigured caps can severely underprice assets, triggering liquidation cascades and TVL loss.

Documentation/Proof:
Approved Proposal

Contracts:
0x7ABB46C690C52E919687D19ebF89C81A6136C1F2


Risk Oracle:
Pendle PT Risk Oracle

Function:
Dynamically determines discount rate for asset pricing and governs liquidation mechanics for PT-assets based on asset maturity
Killswitch to halt lending activity in oracle mispricing / liquidity exhaustion scenarios

Parameters Set:
liquidation threshold, LTV, liquidation bonus , discount rate

Assets Secured:
All PT Assets (PT-sUSDe, PT-USDe) across all Aave instances

Value Secured as of Oct 30, 2025 (% of Aave TVL)*
≈ $2B (3%)

How TVS is Secured (Failure Scenario / Loss Quantification):
Misconfigured liquidation thresholds (i.e., LT=0) would cause mass liquidation of all PT-related assets;

Misconfigured discount rate can lead to oracle mispricing, and (without killswitch) lead to bad debt accumulation

Documentation/Proof:
Approved Governance Proposal

Contracts:
0x7ABB46C690C52E919687D19ebF89C81A6136C1F2


Risk Oracle:
Interest Rate Risk Oracle (WETH)

Function:
Dynamically adjusts the borrowing cost of WETH on Prime

Parameters Set:
slope1

Assets Secured:
WETH (Prime Instance)

Value Secured as of Oct 30, 2025 (% of Aave TVL)*:
≈ $330M

How TVS is Secured (Failure Scenario / Loss Quantification):
Misconfigured slope1 can push borrowing costs to the extreme, forcing mass unwinding of positions

Documentation/Proof:
Approved Governance Proposal

Contracts:
0x7ABB46C690C52E919687D19ebF89C81A6136C1F2


Industry Recognition

The relevance and necessity of Risk Oracles have been recognized widely across the ecosystem. Core ecosystem participants like ACI, BGD Labs, and others have publicly endorsed their importance in safeguarding protocol solvency and user assets. Here are a few highlights:


Conclusion

We believe this reaffirms the importance of risk oracles as it relates to how protocols secure value today. The precedent is clear from cases like GMX and Jupiter, where Chaos Risk Oracles have already been deployed and have secured protocol value for almost a year. These systems actively protect billions in user assets, and their reliability is critical to maintaining protocol stability and preventing liquidations or bad debt.

We’re happy to discuss the details further, including how we measure secured value, the governance parameters managed, and current integrations. We can also share the adapter PR or any supporting data and transaction hashes you’d like to review.

– Chaos Labs

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant