The Eclipse Foundation has introduced ADL (Agent Definition Language) functionality to its LMOS (Language Models Operating System) AI project. The announcement came on October 28.

The goal of Eclipse LMOS is to create an open platform where AI agents can be developed and integrated across networks and ecosystems, according to Eclipse. Built on standards like Kubernetes, LMOS is in production with Deutsche Telekom, one of the largest enterprise agentic AI deployments in Europe, Eclipse said. ADL, meanwhile, addresses the complexity of traditional prompt engineering. ADL provides a structured, model-agnostic framework enabling engineering teams and businesses to co-define agent behavior in a consistent, maintainable way, Eclipse said. This shared language increases the reliability and scalability of agentic use cases; enterprises can design and govern complex agentic systems with confidence, according to Eclipse. This capability further distinguishes Eclipse LMOS from proprietary alternatives, Eclipse added. “With Eclipse LMOS and ADL, we’re delivering a powerful, open platform that any organization can use to build scalable, intelligent, and transparent agentic systems,” Eclipse Executive Director Mike Milinkovich said in a statement.

Eclipse LMOS is designed to let enterprise IT teams leverage existing infrastructure, skills, and devops practices, Eclipse said. Running on technologies including Kubernetes, Istio, and JVM-based applications, LMOS integrates into enterprise environments, accelerating adoption while protecting prior investments. ADL’s introduction empowers non-technical users to shape agent behavior; business domain agents, instead of just engineers, can directly encode requirements into agents, accelerating time-to-market and ensuring that agent behavior accurately reflects real-world domain knowledge, the foundation said. In addition to Eclipse LMOS ADL, LMOS is composed of two other core components:

• Eclipse LMOS ARC Agent Framework: A JVM-native framework with a Kotlin runtime for developing, testing, and extending AI agents comes with 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, currently in the alpha state of development.