Now accepting beta users

Control
Everything

The AI desktop agent that gets things done — private, accessible, and ready out of the box.

Read the Vision
⌘

Greatness, how was your night?

How can I assist you with your desktop tasks today?

Wake word detected

Features.

Powerful local-first capabilities with dedicated Ask & Act modes.

A supercharged knowledge assistant. Analyze PDF documents, ask complex questions, or get help with coding. It processes local files securely without uploading them.

An autonomous agent that controls your mouse and keyboard. "Organize my downloads," "Send this email," or "Play my focus playlist." It sees your screen and gets it done.

Native support for Gemini, OpenRouter (Claude, GPT-4), and local Ollama models for total autonomy.

Control everything with Ctrl+Space, Alt+Z, and custom shortcuts designed for speed.

4-digit SHA-256 encrypted PIN to secure your local assistant from unauthorized access.

Hands-free operation with ambient wake-word detection for a truly hands-off experience.

Local processing by default. Your data and interactions stay strictly on your device.

Hidden from screen sharing and screenshots, ensuring your privacy while you work.

Accessibility is key.

Your workspace, your rules.

Settings

C
Greatness
grucookorie08@gmail.com
Ready
Appearance
Theme
Choose between light and dark mode
Border Streak Effect
Show purple light streak along window borders
Security
Enable Security PIN
Require PIN to open chat
Change PIN
Update your security PIN
Voice & Speech
Voice Activation
Wake word: "hey control"
Auto-Send After Wake Word
Automatically send transcriptions after wakeword
Wake Word Toggle Chat
Allow "Hey Control" to toggle chat open/close
Voice Response
Read AI responses aloud
Mute Notifications
Disable notification sounds
Window Visibility
Allow windows in screenshots/recordings
Greeting Voice
Speak welcome greeting aloud
Edge Glow Effect
Show purple edge glow during Act mode
Proceed Without Confirmation
Allow AI to perform high-risk tasks automatically
AI Model Settings
Model Provider
Choose where to run your AI models
Gemini Model
Default native model: gemini-2.5-flash
ACTIVE (Google SDK)
Shortcuts
Toggle Chat
Show/hide chat window
Ctrl+Space
Stop Task
Emergency stop AI action
Alt+Z
Reset Hotkeys
Restore default shortcuts
System
Run on Startup
Automatically start Control when computer turns on
Floating Button
Show overlay floating button
Lock App
Lock the application
Log Out
Sign out of your account
Quit Application
Exit Control

Join the Beta.

Be among the first to experience the future of local-first AI. Early testers get lifetime pro access.

What can Control do?

Control is simply a situation where a computer-use agent is maximized, it can operate across apps like a power user: read screens, click, type, navigate, and chain actions into real outcomes.

Universal Integration
Works seamlessly across Windows, macOS, and Linux without complex configuration.
Local Autonomy
Processes your data locally, ensuring your privacy remains the top priority.
Voice Precision
Advanced wake-word detection and noise cancellation for reliable hands-free use.
Security First
4-digit PIN protection and SHA-256 encryption keep your personal assistant secure.
AI Versatility
Switch between high-performance cloud models and ultra-private local models on the fly.
Universal Integration
Works seamlessly across Windows, macOS, and Linux without complex configuration.
Local Autonomy
Processes your data locally, ensuring your privacy remains the top priority.
Voice Precision
Advanced wake-word detection and noise cancellation for reliable hands-free use.
Security First
4-digit PIN protection and SHA-256 encryption keep your personal assistant secure.
AI Versatility
Switch between high-performance cloud models and ultra-private local models on the fly.

Frequently Asked Questions

Find answers to common questions about Control

General

Technical

Privacy & Security

Beta Program