Infrastructure Foundation for Agent Interoperability

AGNTCY tackles critical infrastructure challenges that have hindered the seamless collaboration of AI agents in large-scale enterprise environments. The framework delivers essential functionalities akin to traditional networking protocols, but tailored specifically for autonomous software agents.

Agent discovery is powered by the Open Agent Schema Framework, which acts as a decentralized directory allowing agents to broadcast their functions and find specialized services across organizational borders. This supports dynamic, on-the-fly agent collaboration instead of rigid, preconfigured integrations. For example, a customer support agent dealing with a complex issue can autonomously identify and route tasks to dedicated agents for billing, technical troubleshooting, or account updates—without human intervention.

Trust and identity are managed through a cryptographic credential system that enables secure interactions between agents across disparate security zones. Moving beyond conventional API keys or shared secrets, this identity layer supports delegation chains, letting human users securely authorize agents to act on their behalf. This is crucial in regulated sectors where compliance demands clear audit trails linking automated actions to responsible individuals.

At the heart of AGNTCY’s communication system is the Secure Low-Latency Interactive Messaging (SLIM) protocol, designed to handle diverse interaction models such as request-response, publish-subscribe, and persistent bidirectional streams. SLIM incorporates quantum-resistant encryption and sustains extended, multi-phase conversations between agents that may last over days.

Comprehensive observability tools offer full transparency into agent-driven workflows, monitoring not only system performance but also decision logic and resource consumption across interconnected agent networks. This visibility is vital for diagnosing issues in complex, interdependent agent ecosystems where failures can propagate unpredictably.

Synergy with the Agent2Agent Protocol

Cisco’s contribution of AGNTCY to the Linux Foundation complements recent developments in agent communication standards. The Agent2Agent (A2A) protocol, contributed to the same foundation in June, defines standardized message formats and interaction patterns for direct agent communication. AGNTCY enhances A2A by delivering the underlying infrastructure needed for large-scale agent discovery, trust management, and operational oversight.

ForbesKey Tech Firms Unite As Google Donates A2A To Linux FoundationBy Janakiram MSVThis synergy mirrors the layered design of the Internet, where transport protocols like TCP/IP coexist with DNS for service discovery. AGNTCY’s directory services enable A2A-equipped agents to be located and accessed across enterprise networks, while SLIM ensures secure, low-latency transmission of A2A messages with enterprise-grade reliability.

Additionally, AGNTCY seamlessly works with Anthropic’s Model Context Protocol, allowing agents to exchange contextual data and tools via AGNTCY’s secure messaging backbone. This multi-standard integration reduces dependency on any single vendor and empowers organizations to integrate diverse agent technologies into a unified ecosystem.

Market Movement and Industry Trajectory

AGNTCY has garnered strong industry support, with founding contributors including Dell Technologies, Google Cloud, Oracle, Red Hat, and Cisco. This coalition underscores a growing consensus that scalable agent interoperability requires open, vendor-neutral governance.

Dell’s involvement highlights the need for agent coordination across hybrid environments—from cloud data centers to edge computing nodes. Oracle’s participation focuses on strengthening identity verification and data governance in complex, distributed architectures. Google Cloud’s engagement builds upon its earlier A2A initiative, expanding its scope into broader infrastructure orchestration.

By placing AGNTCY under the Linux Foundation, Cisco signals a pivotal shift—from experimental AI agents to foundational enterprise infrastructure. The open, neutral stewardship model mitigates vendor lock-in risks and provides the long-term stability necessary for strategic technology investments in multi-agent systems.

The above is the detailed content of Cisco Donates AGNTCY To Linux Foundation To Advance Multi-Agent Systems. For more information, please follow other related articles on the PHP Chinese website!