The Eclipse Foundation has announced the release of an open-source framework designed to standardize the way companies build and deploy AI agents across their operations. The new Agent Definition Language (ADL) was developed to address a growing challenge in enterprise technology: the lack of common standards for creating AI systems capable of performing complex business tasks autonomously.
The new Eclipse Language Model Operating System (LMOS) is an open-source platform comprising three components:
Eclipse LMOS ADL (Agent Definition Language): A structured, model-neutral language and visual toolkit that lets domain experts define agent behavior reliably and collaborate seamlessly with engineers.
Eclipse LMOS ARC Agent Framework: A JVM-native framework with a Kotlin runtime for developing, testing, and extending AI agents, featuring a built-in visual interface for quick iterations and debugging.
Eclipse LMOS Platform: An open, vendor-neutral orchestration layer for agent lifecycle management, discovery, semantic routing, and observability, built on the CNCF stack and currently in Alpha.
"Agentic AI is redefining enterprise software, yet until now there has been no open source alternative to proprietary offerings," said Mike Milinkovich, executive director of the Eclipse Foundation, in a statement. "With Eclipse LMOS and ADL, we're delivering a powerful, open platform that any organisation can use to build scalable, intelligent, and transparent agentic systems."
The foundation is positioning ADL as vendor-neutral, built on open standards including Kubernetes and Alpha, with operations running on cloud infrastructure. The system aims to allow AI agents to function across different networks and technology ecosystems.
The business case for such standardization is significant. The Eclipse Foundation cited Gartner research projecting that by 2028, 15% of daily business decisions will be made autonomously by AI agents, rising to 33% by 2033.
Traditional methods of designing AI agent behavior through prompt engineering have proven complex and difficult to scale, according to the foundation. ADL attempts to solve this by providing a structured framework that can be maintained and updated more systematically than custom-built alternatives.
"With ADL, we wanted to make defining agent behaviour as intuitive as describing a business process, while retaining the rigor engineers expect," said Arun Joseph, Eclipse LMOS project lead, in a statement. "It eliminates the fragility of prompt-based design and gives enterprises a practical path to scale agentic AI using their existing teams and resources."
The platform is designed to separate the technical infrastructure from business logic, potentially allowing non-technical teams to participate in defining agent capabilities while engineers handle implementation details.