Phase 2: Prime Brokerage
Theory:
We believe that the key to expanding undercollateralized credit in DeFi post Phase 1 lies in creating a "borrowing in a bubble" environment.
This concept allows us to simulate undercollateralized credit while maintaining technical overcollateralization by retaining ultimate control of the assets.
In traditional finance, prime brokers provide leverage, market access, and other benefits to clients while executing trades and transactions on their behalf. We can replicate this model on-chain by defining clear boundaries and executing transactions within a controlled environment.
The core idea is to issue funds to a restricted account that only allows a well-defined set of commands and smart contract interactions. This approach ensures that borrowed funds can be put to productive use & funds cannot be freely transferred to an unrestricted, self-custodial wallet
Algorithmic Security:
By implementing this "algorithmic security," we create a system where:
You can't default, because the code won't let you
The protocol retains control while giving users significant flexibility
Risk is contained within predefined parameters
This model allows us to offer higher LTV's => hence higher leverage ratios => hence borrowers get increased capital efficiency while maintaining the security standards necessary in a trustless environment.
Key Components:
Credit Accounts: Smart contract wallets that hold both user collateral and borrowed funds, with restrictions on withdrawals and interactions.
Whitelisted Protocols: A carefully curated list of Defi protocols and assets that Credit Accounts can interact with, balancing opportunity with risk management.
Risk Assessment Engine: A sophisticated system for evaluating the risk of various assets and protocol interactions in real-time.
Liquidation Mechanisms: Automated processes to ensure positions remain overcollateralized from the protocol's perspective, even as users access undercollateralized leverage.
Technical Overview:
Refer to this section -> What goes underneath the hood?
Usecases
Refer to this section -> What can you do?
Last updated