The Linux Foundation is accelerating its push into agentic AI, unveiling a third major project in as many months aimed at creating secure, interoperable infrastructure for autonomous agents.
Following the adoption of two recent agentic AI projects in June and July, LF added the Agent Gateway Project on August 25 at the Open Source Summit EU. Created by the cloud-native application networking company Solo.io, it optimizes connectivity, security, and observability in agent-based AI environments.
Numerous gateways exist, but most were designed before the rise of AI agents and struggle to support modern protocols without significant architectural changes, according to LF. To keep pace with the evolving artificial intelligence (AI) industry, organizations need a gateway specifically designed for today’s dynamic ecosystem.
LF began supporting mass innovation through open-source solutions at June’s Open Source Summit North America, where it announced the launch of the Agent2Agent (A2A) project. The A2A initiative is based on an open protocol Google launched in April for secure agent-to-agent communication and collaboration.
The growing support within LF for agentic AI intensified in late July when LF welcomed the Agntcy project. This open-source infrastructure enables discovery, identity, messaging, and observability among AI agents from different vendors and frameworks. It allows dynamic multi-agent environments by making A2A agents and Model Context Protocol (MCP) servers discoverable through Agntcy directories.
“Agentic AI demands purpose-built infrastructure, not just another software layer,” Jon Alexander, senior vice president, cloud technology at Akamai Technologies, told LinuxInsider.
Meeting this demand requires rethinking compute, storage, and data movement from the ground up, making retrofitting legacy systems not feasible.
Innovating Security Solutions
AgentGateway is the first and only data plane built from the ground up for AI agents, governing and securing communication across agent-to-agent, agent-to-tool, and agent-to-LLM interactions, LF noted. It supports leading protocols, including A2A, which was recently contributed to the foundation, and Anthropic’s MCP.
With LF’s neutral governance, the Agent Gateway Project will remain vendor-agnostic and community-driven, supporting collaboration, intellectual property management, and long-term stewardship.
According to Alexander, the Agent Gateway Project represents a solid step toward safeguarding the future of agentic AI.
“We’re happy to see it hosted by The Linux Foundation, where open source and community can drive the adoption and longevity that AI infrastructure requires,” he said.
Justin Cappos, professor at New York University and creator of the The Update Framework (TUF), Uptane, and in-toto projects, agreed that the open-source community has yet to solve many of the problems resulting from AI’s rapid development.
“One of the biggest open security problems today is how to do MCP security effectively,” he told LinuxInsider.
Tripling Down on Innovation
Numerous problems converge in this space, many of which are beyond the community’s current ability to address. The Agent Gateway Project provides a first step toward addressing critical issues with basic role-based access control and visibility of actions to MCP servers.
Google developed the protocol 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.
According to Rao Surapaneni, VP and GM of business applications Platform for Google Cloud, it addresses the growing need for agents to operate in dynamic, multi-agent environments, coordinating actions across diverse apps and data infrastructures.
With it, autonomous agents can discover one another, exchange information securely, and collaborate across systems. This capability allows developers and organizations to consolidate agents from multiple sources and platforms, enhancing modularity, mitigating vendor lock-in, and accelerating innovation.
“The Agent2Agent protocol establishes a vital open standard for communication, enabling the industry to build truly interoperable AI agents across diverse platforms and systems,” he said.
Delivering Critical Solutions
The Agntcy project is foundational to upgrading security with agentic AI applications. It is interoperable with leading AI agent technologies, including the A2A project.
“The Agntcy project lays groundwork for secure, interoperable collaboration among autonomous agents,” said Jim Zemlin, executive director of The Linux Foundation. “We are pleased to welcome the Agntcy project to the Linux Foundation to ensure its infrastructure remains open, neutral, and community-driven.”
John Roese, global CTO and chief AI officer at Dell Technologies, agreed that the project is essential. Dell has been building Agntcy’s evaluation and observability components from day one because reliable AI agents cannot scale without purpose-built monitoring.
“Moving all components of Agntcy to The Linux Foundation ensures these tools serve the entire ecosystem, not just our customers. As a founding member of Agntcy, we’re eager to see neutral governance accelerate adoption of standards we know enterprises need for production agent deployments,” he said.