Recently, the Thier Terminal Laboratory of the China Academy of Information and Communications Technology (CAICT), together with the International Internet Finance Authentication Alliance (IIFAA) and more than twenty leading companies, officially released the "Agent Security Link (ASL)" protocol.

With the rapid development of intelligent agent technology, interconnect communication protocols such as MCP and A2A have emerged one after another, but protocol collaboration still faces three major trust challenges: First, protocol links are untrustworthy, collaborator identities are difficult to verify, communication links are easily spoofed, and communication data is easily tampered with; Second, intent transmission distortion, meaning intent can be tampered with or replaced during cross-subject transmission, and execution deviates from the user's original purpose; Third, authorization is out of control, making it difficult to express, test, and trace the boundaries of authorization in multi-level collaboration. To address these issues, CTTL Terminal Labs of CAICT's has jointly developed the ASL protocol framework for trusted interconnection of general agents, complementing MCP, A2A, and other protocols, covering four core capabilities: trusted identity, trusted connection, trusted intent, and trusted authorization, providing secure and trustworthy collaboration guarantees for agent interconnection. Additionally, ASL focuses on the core security needs of agent collaboration, providing systematic support for various business scenarios, including instant task execution, scheduled task execution, and multi-level delegated task execution.

Next, the CTTL Terminal Labs of CAICT will continue to improve the construction of safe and trustworthy interconnection standards for agents, collaborate with industry enterprises to develop products and services according to relevant standards, promote the establishment of an evaluation index system for trusted interconnection protocols for agents, further enhance the standardization of agents, and facilitate their healthy and orderly development.

Contact Person:

MS. Wei

15611638262

weifanxing@caict.ac.cn