A macOS app that blurs your screen when you slouch.
Posturr uses your Mac's camera and Apple's Vision framework to monitor your posture in real-time. When it detects that you're slouching, it progressively blurs your screen to remind you to sit up straight. Maintain good posture, and the blur clears instantly.
- Real-time posture detection - Uses Apple's Vision framework for body pose and face tracking
- Progressive screen blur - Gentle visual reminder that intensifies with worse posture
- Multi-display support - Works across all connected monitors
- Privacy-focused - All processing happens locally on your Mac
- Lightweight - Runs as a background app with minimal resource usage
- No account required - No signup, no cloud, no tracking
- Download the latest
Posturr-vX.X.X.zipfrom the Releases page - Unzip the downloaded file
- Drag
Posturr.appto your Applications folder
Since Posturr is not signed with an Apple Developer certificate, macOS Gatekeeper will initially block it:
- Right-click (or Control-click) on
Posturr.app - Select "Open" from the context menu
- Click "Open" in the dialog that appears
- Grant camera access when prompted
You only need to do this once. After the first launch, you can open Posturr normally.
Posturr requires camera access to monitor your posture. When you first launch the app, macOS will ask for permission. Click "OK" to grant access.
If you accidentally denied permission, you can grant it later:
- Open System Settings > Privacy & Security > Camera
- Find Posturr and enable the toggle
Once launched, Posturr runs in the background and displays a brief "Claude Mode Active" notification. The app:
- Continuously monitors your posture through the camera
- Applies screen blur when poor posture is detected
- Automatically clears the blur when you correct your posture
- Press Escape anywhere to quit the app
- Position your camera at eye level when possible
- Ensure adequate lighting on your face
- Sit at a consistent distance from your screen
- The app works best when your shoulders are visible
Posturr uses Apple's Vision framework to detect body pose landmarks:
- Body Pose Detection: Tracks nose, shoulders, and their relative positions
- Face Detection Fallback: When full body isn't visible, tracks face position
- Posture Analysis: Measures the vertical distance between nose and shoulders
- Blur Response: Applies screen blur proportional to posture deviation
The screen blur uses macOS's private CoreGraphics API for efficient, system-level blur that covers all windows and displays.
- macOS 13.0 (Ventura) or later
- Xcode Command Line Tools (
xcode-select --install)
git clone https://github.com/yourusername/posturr.git
cd posturr
./build.shThe built app will be in build/Posturr.app.
# Standard build
./build.sh
# Build with release archive (.zip)
./build.sh --releaseswiftc -O \
-framework AppKit \
-framework AVFoundation \
-framework Vision \
-framework CoreImage \
-o Posturr \
main.swift- No code signing: Requires manual Gatekeeper bypass on first launch
- Camera dependency: Requires a working camera with adequate lighting
- Detection accuracy: Works best with clear view of upper body/face
- No preferences UI: Currently no graphical settings interface
Posturr exposes a file-based command interface for external control:
| Command | Description |
|---|---|
capture |
Take a photo and analyze pose |
blur <0-64> |
Set blur level manually |
quit |
Exit the application |
Write commands to /tmp/posturr-command. Responses appear in /tmp/posturr-response.
- macOS 13.0 (Ventura) or later
- Camera (built-in or external)
- Approximately 10MB disk space
Posturr processes all video data locally on your Mac. No images or data are ever sent to external servers. The camera feed is used solely for posture detection and is never stored or transmitted.
MIT License - see LICENSE for details.
Contributions are welcome! Please feel free to submit issues and pull requests.
- Built with Apple's Vision framework for body pose detection
- Uses private CoreGraphics API for efficient screen blur
- Inspired by the need for better posture during long coding sessions