Cobo proposes the "Contracted Permissions + MPC" framework: building a controllable trust layer for the Agent economy

Mars Finance News, April 21 — Cobo AI Growth Lead Brad Bao shared at the offline event themed “Decoding Web 4.0: When AI Agents Take Over On-Chain Permissions” that Cobo is reconstructing the fund control logic of Agents through a structured permission system. In their product, the relationship between users and Agents is no longer a simple wallet authorization but is defined by an executable “contract (Policy/Agreement)” that sets permission boundaries.
This contract at least includes four core elements: Intent, Path, Rules, and Termination conditions, thereby upgrading the traditional “all or nothing” permission control to a fine-grained governance mechanism that is negotiable, auditable, and revocable.
At the execution level, Cobo combines MPC (Multi-Party Computation) technology to ensure that even if the Agent exhibits abnormal behavior, the system is attacked, or prompt injection risks occur, the funds remain under user control, and no single entity can bypass constraints to transfer assets independently.
Brad Bao emphasized that under this architecture, Agents can gain enough operational freedom to complete complex tasks, but their actions are always confined within verifiable boundaries.
He stated that this system essentially builds a “trust and insurance infrastructure” within the Agent economy, combining “contracted permissions + secure computation” to achieve controllable delegation of Agent behavior, ultimately improving automation efficiency while safeguarding user assets and ensuring outcome certainty.