Ora AI Intelligence Suite
A high-performance Windows 11 desktop automation agent and cloud intelligence suite powered by Gemini 2.0, GPT-4o, local Ollama, Electron, and Python sidecar OS control.
Autonomous Computer Use & Cloud Intelligence.
Extracted directly from the Ora AI repository specs, here is what powers the Ora desktop automation platform.
Real OS Automation Engine
PyAutoGUI + pywin32 integration
Executes real OS-level mouse movements, clicks, and keyboard inputs using a high-performance Python sidecar process. Features visual AI cursor feedback showing action targets in real time.
Multi-Model AI Gateway
Gemini 2.0, GPT-4o & Local Ollama
Flexible multi-model AI routing supporting Google Gemini 2.0 Flash Vision (with Computer Use), OpenAI GPT-4o, and local Ollama inference on localhost for 100% offline privacy.
Global Hotkey & Overlay
Alt+Space command bar
Summon or dismiss Ora AI instantly anywhere in Windows 11 with the global Alt+Space hotkey. Features a sleek Fluent UI overlay with Mica & Acrylic transparency effects.
High-Speed Screen Capture
Native mss library capture
Captures full-screen desktop frames at 60 FPS using Python's native mss library, analyzing visual elements before dispatching precise click coordinates to the automation process.
Zustand State & IPC Bridge
Electron context bridge
Type-safe IPC communication channel connecting Electron main process, React frontend, and Python sidecar via JSON stdin/stdout streams with Zustand state management.
Action History & Logging
Audit trail & rollback
Maintains a real-time visual history of all AI actions, executed key combinations, mouse coordinates, and model reasoning steps for full transparency and safety.
Built for local speed and multi-model flexibility.
Ora AI combines Electron, React 19, Next.js 16, Python OS sidecars, and local LLM inference.
Electron + React 19
Cross-platform desktop runner with context isolation
Next.js 16 + Tailwind CSS v4
Sleek web companion and dashboard surface
Python Sidecar (PyAutoGUI)
Low-level OS process handling mouse, keyboard, and screen capture
Gemini 2.0 / GPT-4o / Ollama
Multi-provider AI routing for cloud & offline execution
Zustand + Electron IPC
Bi-directional JSON stream protocol
Fluent UI + Glassmorphism
Windows 11 native Mica, Acrylic, and blur effects