This project, encompassing a custom MCP server written in Dart, acts as an AI-powered IT assistant that integrates the Model Context Protocol (MCP) into real-world IT operations. It runs a local MCP server, connecting it with Gemini 2.5 Pro to enable natural language control over a wide range of IT tasks, including file management, document editing, cloud service queries (Azure AD, Intune, SharePoint, OneDrive), GitHub exploration, Windows Server management, and Google services (Gmail, Calendar). The unique Dart-based server offers seamless Flutter integration, lightweight performance, strong typing, and a unified development stack, providing a developer-friendly architecture for AI as a unified control layer across local systems and cloud platforms.
Key Features
01AI-powered natural language control for IT tasks via Gemini 2.5 Pro
02Reusable MCP server component compatible with other MCP-enabled clients
03Lightweight and fast startup ideal for local subprocess execution
04Comprehensive IT task automation: file management, document editing, cloud services, server monitoring
050 GitHub stars
06Unified Dart/Flutter architecture for both UI (client app) and backend (MCP server)
Use Cases
01Automating routine file and folder operations, including intelligent organization, using plain English commands.
02Querying and managing cloud resources (Azure AD, Intune, SharePoint, OneDrive) and developer platforms (GitHub) through natural language.
03Reading, searching, editing, and generating Excel/Word documents and managing emails/calendar events without opening applications.
