Aave Labs launches AI-driven governance security system Aave Checkpoint

Aave Labs announced the launch of the AI-driven governance security system Aave Checkpoint, which adds a structured multi-layer review process for each proposal and load before execution.

It is reported that the workflow of Checkpoint is divided into two phases: the automatic AI analysis phase, where the system automatically retrieves on-chain data, proposal source code, and IPFS text when a new proposal is submitted, cross-references the parameters with governance specifications, uses Seatbelt to simulate output and check the proposal execution path, runs over 60 specialized security analysis agents for vulnerability detection, DeFi-specific risk analysis, agent upgrade security checks, etc., and generates a structured audit report with pass/fail conclusions; and the mandatory manual review phase, where each AI-generated report must be independently signed off by at least two people before it can be considered reviewed. Checkpoint covers Aave V3, V4, GHO, and Aptos-v3 protocols, including 5 custom Aave governance skills and 52 external security skills.