Amazing-Marvin

Amazing Marvin AI Assistant Integration connects your Amazing Marvin productivity data with AI assistants to provide personalized task management help. The server exposes tools that allow your AI to view and interact with your real Marvin data, including daily overviews, project insight, task creation and completion, as well as time tracking. When you ask your AI questions like what to focus on today or show me my projects, the MCP routes these requests to specific tools such as get_daily_productivity_overview(), get_tasks(), get_projects(), get_due_items(), create_task(), mark_task_done(), and start_time_tracking(), enabling the AI to generate context-aware recommendations and actions based on your current workload, deadlines, and priorities. No manual data copying is required; your AI simply queries the live Marvin data through the MCP interface.

To use it, install the Python package and run the server in your environment (or connect via your MCP client). The provided tools are designed to work with your Marvin setup (work vs. personal projects, categories, labels, due dates, priorities, and statuses). Your AI will automatically map your Marvin structure so you can ask for daily planning, project progress, overdue items, or time-tracking actions, and receive precise, actionable results tailored to your real data.

Prerequisites:\n- Python 3.10+ (as indicated by the project)\n- An Amazing Marvin account with API access and your API key\n- An MCP-compatible AI client (Claude, Cursor, Windsurf, VS Code, etc.) or a direct Python environment\n\nInstallation steps:\n1) Create a virtual environment (optional but recommended):\nbash\npython -m venv venv\nsource venv/bin/activate # macOS/Linux\nvenv\Scripts\activate # Windows\n\n2) Install the MCP package from PyPI:\nbash\npip install amazing-marvin-mcp\n\n3) Configure your environment with your API key (example shown in the config snippet):\nbash\nexport AMAZING_MARVIN_API_KEY=your-api-key-here # macOS/Linux\nset AMAZING_MARVIN_API_KEY=your-api-key-here # Windows\n\n4) Run the MCP server using Python: (as shown in the mcp_config)\nbash\npython -m amazing_marvin_mcp\n\n5) If you plan to connect via an MCP client, ensure the client’s configuration points to the server and environment variable is accessible to the process.

Environment variables and configuration tips:\n- The API key for Amazing Marvin must be provided as AMAZIN_G_MARVIN_API_KEY in your environment or via the MCP client config, as demonstrated in the example config. Without this, the server cannot access your Marvin data.\n- In client configurations (Claude Desktop, Cursor, Windsurf, VS Code, etc.), the mcpServers entry should include the command and arguments to invoke the Python module ( -m amazing_marvin_mcp ) and the required environment variable.\n- If you encounter API key not found or connection issues, double-check that the API key is correctly set, no leading/trailing spaces exist, and the process has access to the environment variable.\n- The MCP server supports common productivity queries such as daily overviews, project views, due items, and time tracking; you can extend usage by enabling additional Marvin API capabilities as needed.\n- This integration emphasizes privacy by design, keeping Marvin data within your controlled environment and only exposing what your AI queries through the MCP tools.