![]() |
VOOZH | about |
We’re so glad you’re here. You can expect all the best TNS content to arrive Monday through Friday to keep you on top of the news and at the top of your game.
Check your inbox for a confirmation email where you can adjust your preferences and even join additional groups.
Follow TNS on your favorite social media networks.
Become a TNS follower on LinkedIn.
Check out the latest featured and trending stories while you wait for your first TNS newsletter.
Cisco today announced that it is donating the AGNTCY project — an open source infrastructure project that aims to enable discovery, identity, messaging and observability to multiagent systems — to the Linux Foundation.
The project comes out of Cisco’s in-house Outshift incubator and was open sourced in March 2025. It includes specs, code and services, ranging from a discovery layer that is akin to a DNS system for agent communication and an agent directory, to a tool for verifying an agent’s identity, an observability framework and a protocol to invoke and configure remote agents.
Dell, Google Cloud, Oracle and Red Hat are joining Cisco as founding members under the Linux Foundation umbrella. Galileo, LangChain, MongoDB, Traceloop, Weaviate and about 70 additional companies that have supported the project since launch will join the project at the Linux Foundation, too.
It’s worth noting that AGNTCY isn’t meant as a replacement for protocols like Google’s Agent2Agent (A2A) — which Google recently donated to the Linux Foundation as well — or Anthropic’s Model Context Protocol (MCP), which isn’t part of the Linux Foundation (yet).
Instead, this project is about the rest of the infrastructure needed for agents to securely and reliably communicate with each other and third-party services.
“Open, community-driven standards are essential for creating a diverse, interoperable agentic AI ecosystem,” said Rao Surapaneni, VP/GM, business application platform, Google Cloud. “We’re pleased that Cisco is moving AGNTCY to the Linux Foundation, where it will be neutrally governed alongside the Agent2Agent protocol to advance powerful, collaborative agent systems for the industry.”
These open standards are important, the AGNTCY backers argue, to ensure that the AI agent ecosystem doesn’t fragment into different vendor silos that will make it harder for them to communicate and collaborate across these platforms.
“Think about AGNTCY as a full life cycle problem, where what we are enabling is a distributed, agentic cloud architecture,” Vijoy Pandey, SVP/GM of Outshift by Cisco, told me when the project first launched.
“The Internet enabled e-business at scale. And then the world went towards the cloud, and the cloud internet, and that enabled mobile-first applications at scale, and social interactions at scale, and SaaS [Software as a Service] at scale. And now we feel that we need to connect agentic endpoints, and that’s what we’re calling the internet of agents. But it goes a step further, and basically what we want to do is enable distributed agentic competition.”
Specifically, the AGNTCY framework includes the Open Agent Schema Framework, which the group would like to become the standard for defining an agent’s capabilities and dependencies, which works hand in hand with the project’s directory to offer a DNS-like service for agent discovery.
There is also an agent identity service, which allows agents to prove who they are and that they are authorized to perform certain actions, as well as a messaging protocol (SLIM, for Secure Low-latency Interactive Messaging) for enabling low-latency communication between agents, tools and humans. AGNTCY also includes an observability framework and data schema to provide additional visibility into how agents work across different systems.