TLDR: The Linux Foundation has announced a significant initiative to foster seamless collaboration among disparate AI agents by welcoming Cisco’s AGNTCY project under its open-source governance. This move aims to establish an ‘Internet of Agents,’ preventing fragmentation and enabling secure, cross-platform communication for AI programs.
In a pivotal move to address the growing challenge of interoperability within the rapidly expanding field of artificial intelligence, the Linux Foundation has announced its commitment to fostering a unified ecosystem for AI agents. On July 29, 2025, the non-profit organization officially welcomed the AGNTCY (pronounced ‘agency’) project, an open-source infrastructure initially developed and open-sourced by Cisco in March 2025. This initiative is designed to create a foundational, open-standard infrastructure, envisioned as an ‘Internet of Agents,’ that will enable AI agents from various vendors and frameworks to communicate and collaborate seamlessly.
The proliferation of specialized AI agents has led to concerns about fragmentation and the creation of isolated silos, hindering their collective utility. The AGNTCY project directly tackles this issue by providing core components for AI agent discovery, secure messaging, and cross-platform collaboration. It aims to deliver secure agent identity, reliable messaging, and end-to-end observability, enhancing transparency, performance, efficiency, and trust within multi-agent systems.
The project has garnered substantial industry backing, with over 75 companies supporting it since its initial open-sourcing. Formative members under Linux Foundation governance include major players such as Cisco, Dell Technologies, Google Cloud, Oracle, and Red Hat. This broad industry support underscores the critical need for a universal standard to prevent a fragmented AI landscape, reminiscent of past technology battles over standards.
Vijoy Pandey, GM and SVP of Outshift by Cisco, emphasized the importance of community ownership in this endeavor, stating, ‘Building the foundational infrastructure for the Internet of Agents requires community ownership, not vendor control. 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.’ The Linux Foundation’s neutral governance is expected to be instrumental in achieving this widespread adoption and making AGNTCY an industry-wide standard.
Also Read:
- SoftBank Unveils “Agent Firewall” for Secure Cross-Organizational AI Collaboration
- Cisco Fortifies Data Centers Against Emerging AI Agent Threats
AGNTCY is also designed to be interoperable with other leading AI agent technologies, including the Agent2Agent (A2A) project, which was also recently contributed to the Linux Foundation, and Anthropic’s Model Context Protocol (MCP). This ensures that A2A agents and MCP servers can be discovered through AGNTCY directories, and message transport can occur over the Secure Low Latency Interactive Messaging (SLIM) protocol, further solidifying the vision of a truly interconnected ‘Internet of Agents.’ The project is already backed by real-world production use cases, spanning AI-driven CI/CD pipelines, multi-agent IT deployments, and telecom network automation.
