Cisco is donating its AGNTCY initiative to the Linux Foundation, which will continue to advance the AI agent management platform as an open-source project. Outshift, which is Cisco’s research and development arm, launched AGNTCY to develop AI agent discovery, identity, messaging, and observability infrastructure.
After the handover, Cisco, Dell, Google Cloud, Oracle, Red Hat and more than 65 other vendors will build a wide range of industry-standard protocols and frameworks to allow AI agents to discover one another, communicate, collaborate and be managed across platforms, models, and organizations, according to the Linux Foundation and Cisco.
“From IT deployments that coordinate ServiceNow tickets, Cisco networks, and Salesforce customer data to drug discovery pipelines that connect protein modeling agents with automated wet-lab robots; enterprises want to agentify work spanning multiple software systems,” wrote Vijoy Pandey, general manager and senior vice president of Outshift by Cisco, in a blog post.
“But to build these collaborative systems, agents need to be able to find each other, verify their identities, and share context without expensive custom integration work. Agentic AI is now at the same inflection point the early internet faced. Brilliant individual systems that can’t talk to each other, where every agent is its own island – until common protocols emerge,” Pandey wrote.
Agentic AI fragmentation has been accelerating since Cisco launched AGNTCY (along with agent orchestration startup LangChain and security and observability vendor Galileo) in March, with every platform building its own discovery systems, identity frameworks, and messaging protocols, according to Pandey. “The missing piece isn’t smarter agents — it’s complete infrastructure that lets any agent work with any other agent, regardless of who built it or where it runs,” Pandey wrote.
AGNTCY is that complete framework that addresses the multi-agent software lifecycle, from build to runtime, Pandey stated.
Specifically, Cisco’s donation includes:
- Agent discovery: This includes Open Agent Schema Framework (OASF), a standard to record agent capabilities and dependencies, plus a decentralized Agent Directory. “Think DNS-like discovery for agents—any agent can discover and understand what other agents can do and include them in agentic workflows to accomplish tasks,” Pandey wrote.
- Agent identity: Cryptographically verifiable identities and tool-based access control for autonomous agents “[enable] agents to prove who they are and perform authorized actions across vendor and organizational boundaries.”
- Agent messaging: “SLIM (Secure Low-latency Interactive Messaging) handles agent, human, and tool communication patterns in low-latency, multi-modal data exchange, human-in-the-loop interactions, and quantum-safe security by design,” Pandey wrote.
- Agent observability: A framework, data schema and SDK “incorporate the probabilistic nature of AI agents and provide end-to-end visibility across multi-agent, multi-vendor, multi-organization workflows.”
- Protocol integration: “We’re making AGNTCY building blocks interoperable with Agent 2 Agent (A2A) and Model Context Protocol (MCP). A2A-enabled agents and MCP servers can be discovered through AGNTCY directories, transport messages over SLIM, and be monitored with AGNTCY observability SDKs,” Pandey wrote.
In June, the Linux Foundation started an Agent2Agent (A2A) project, designed around an open protocol created by Google for secure agent-to-agent communication and collaboration. Developed to address the challenges of scaling AI agents across enterprise environments, A2A empowers developers to build agents that seamlessly interoperate, regardless of platform, vendor or framework, the Foundation stated.
According to the Foundation, the A2A protocol launched by Google includes support from more than 100 technology companies and addresses the growing need for agents to operate in dynamic, multi-agent environments, coordinating actions across a wide array of applications and data infrastructure.
The AGNTCY stack will be further integrated with A2A work, according to Pandey.
“A complete multi-agent system needs all these pieces working together. Discovery without identity is useless. Messaging without observability is blind. Protocols that synergistically work together. The goal is agent collaboration that just works,” Pandey stated. “Building open and decentralized infrastructure is never a one-company job. The Linux Foundation provides the neutral governance that enterprises trust and the sustainability model that keeps critical projects alive.”
Source:: Network World