# JARVIS — Just A Rather Very Intelligent System **Windows Port | Google Workspace Edition** A voice-activated AI assistant powered by FastAPI, Anthropic Claude, Fish Audio TTS, and OpenAI Whisper — integrated with the full Google Workspace suite. ## Quick Start ### 1. Prerequisites - Python 3.11+ - Google Cloud project with OAuth credentials (`credentials.json`) - Anthropic API key - Fish Audio API key ### 2. Install dependencies ```bash python -m venv .venv .venv\Scripts\activate pip install -r requirements.txt ``` ### 3. Configure environment ```bash cp .env.example .env # Edit .env with your API keys ``` ### 4. Google OAuth setup - Place your `credentials.json` from Google Cloud Console in the project root - On first run, a browser window will open for authorization - Token is saved to `token.json` automatically ### 5. Run ```bash uvicorn main:app --reload --host 0.0.0.0 --port 8000 ``` ### 6. Open dashboard Navigate to `http://localhost:8000` in Chrome. ## Project Structure ``` jarvis-windows/ ├── main.py # FastAPI entrypoint ├── requirements.txt ├── .env.example ├── .gitignore ├── core/ │ ├── google_auth.py # Shared Google OAuth handler │ ├── voice_input.py # Whisper STT │ ├── voice_output.py # Fish Audio TTS │ ├── claude_agent.py # Claude API + tool use │ └── websocket_manager.py # WebSocket connection manager ├── capabilities/ │ ├── screen_vision.py # mss + win32gui screen capture │ ├── calendar_access.py # Google Calendar API │ ├── email_access.py # Gmail API (read-only) │ ├── notes_manager.py # Google Keep / gkeepapi │ ├── task_manager.py # Google Tasks API │ ├── file_system.py # os/pathlib file operations │ ├── terminal_control.py # subprocess PowerShell/CMD │ ├── browser_control.py # Selenium Chrome automation │ └── git_manager.py # Git via subprocess ├── utils/ │ ├── audio_utils.py # PyAudio helpers │ └── window_utils.py # win32gui active window └── client/ └── jarvis-dashboard.html # Animated Three.js frontend ``` ## Roadmap See `ROADMAP.md`