Rome

Rome is an agentic operating environment from Rome OS for collaboration between humans and AI agents. It provides tools, workflows, memory, interfaces, policies, and persistence in a guardrailed environment where agents can build harnesses, define standard operating procedures, and orchestrate workflows under human guidance.

Rome supports one-off tasks, scheduled work, long-running follow-through, and installable Rome Apps. An app combines a purpose-built web interface, agent-owned collaborators, reusable skills, lifecycle hooks, typed actions, HTTP APIs, and persistent app-private databases or files. Apps are defined with an app.yaml manifest and use the @rome-os/app-runtime and @rome-os/app-web-sdk packages; Rome can generate and iterate on app or workflow source code from a plain-language request.

The project is distributed as a Docker-based deployment and as a pnpm monorepo with local dashboard, desktop, runtime, SDK, and first-party app components. Rome Cloud provisions a private environment for each guardian and is described as a preview service.