LazyDesk is a local-first AI assistant designed to automate small digital chores. It aims to help users spend less time on tasks like note organization and document summarization. The assistant performs these tasks using a locally running AI model, eliminating the need for manual effort. The core principle is to process information directly on the user's machine.LazyDesk utilizes an open-weight LLM, local inference with Ollama, and a simple application layer. Users provide text, the local model processes it, and the user reviews the generated output. This approach prioritizes privacy by keeping data on the user's device. Open innovation allows for experimentation with different models and customization of workflows.Local processing also reduces dependence on expensive, hosted AI APIs. Users gain more control over their data and the assistant's functionality. Furthermore, LazyDesk can operate offline once set up, ensuring resilience. The project emphasizes giving users more freedom and automating repetitive tasks to free up time for more meaningful work. It is built with the belief that AI should not add another managed service.