Show HN: OculOS – Give AI agents control of your desktop via MCP (github.com)

OculOS has been launched as a groundbreaking tool that allows AI agents to interface with desktop applications via a simple JSON REST API, turning user interface elements into API accessible endpoints. This lightweight daemon interacts with the OS accessibility tree, enabling users to control any desktop app without the need for screenshots, pixel coordinates, or code injections—essentially making anything on-screen a programmable element. This innovation means that AI models like Claude, GPT, or even custom solutions can autonomously manipulate desktop applications, as demonstrated by Claude Code’s ability to autonomously manage Spotify. For the AI/ML community, OculOS signifies a pivotal advancement in desktop automation and AI integration, potentially streamlining workflows and enabling higher levels of task automation previously constrained by traditional GUI interactions. Its technical support for multiple operating systems, including macOS, Windows, and Linux, alongside features like a built-in interactive element tree, inspector, and live WebSocket events, positions OculOS as a versatile tool for developers looking to enhance AI capabilities in real-world applications. With zero dependencies and easy installation, OculOS could redefine how AI systems interact with user interfaces, fostering a new wave of intelligent applications.