The Linux Foundation today announced it has accepted agentgateway, an open source, AI-native proxy created by Solo.io to optimize connectivity, security, and observability in agentic AI environments. Backed by major industry players including AWS, Cisco, Huawei, IBM, Microsoft, Red Hat, Shell, and Zayo, the project is designed to become the standard for secure communication between AI agents, tools, and large language models (LLMs).

Unlike traditional gateways—many of which were built before the rise of AI agents and struggle to support modern protocols—agentgateway is the first data plane built specifically for AI agents. It enables governance and security across agent-to-agent, agent-to-tool, and agent-to-LLM interactions, addressing the scalability and interoperability challenges organizations face as AI ecosystems evolve.

The project already supports leading protocols such as Agent2Agent (A2A), recently contributed to the Linux Foundation, and Anthropic’s Model Context Protocol (MCP). With neutral governance under the Linux Foundation, agentgateway will remain vendor-agnostic and community-driven, ensuring open collaboration, intellectual property management, and long-term stewardship.

“The future of AI depends on open, neutral infrastructure that allows diverse organizations and technologies to work together,” said Jim Zemlin, executive director of the Linux Foundation. “The agentgateway project brings the industry together with a shared foundation for secure, scalable, and transparent agent communication that’s built to last.”

“agentgateway began as a response to a real challenge we faced in managing secure, contextual communication between autonomous agents, which traditional proxies couldn’t handle,” said Idit Levine, CEO of Solo.io. “By contributing this project to the Linux Foundation, we’re opening the door for organizations everywhere to shape the future of agent connectivity that supports security, interoperability, and long-term collaboration.”

The project is now open for contributions.