Slack Huddle Detector

A macOS utility that detects when you're in a Slack huddle (voice/video call) by monitoring system audio state and power assertions.

To the Slack team if you are reading this: Please make this easier to detect m'kay?

Features

• 🎯 Accurate Detection: Monitors multiple system indicators to reliably detect huddle start and end
• 🔄 Auto-Calibration: Learns your baseline state and adapts thresholds
• 📊 Smart End Detection: Three methods to ensure huddle end is properly detected
• 🚀 Real-time Monitoring: Updates every 3 seconds with current status
• 📈 Dynamic Thresholds: Adjusts detection based on peak scores and trends
• 📱 MenuBar Integration: Optional native macOS menubar app for visual status indicator

Requirements

• macOS (tested on macOS 14.x)
• Python 3.6+
• Slack desktop app
• sudo access (for audio system monitoring)
• Swift 5.9+ (for menubar app)

Installation

1. Clone or download the repository:

git clone https://github.com/yourusername/slack-huddle-detection.git
cd slack-huddle-detection

2. No additional Python packages required - uses only standard library

Usage

Option 1: Python Detector Only

Run the detector with sudo privileges:

sudo python3 slack-huddle-detector-optimized.py

Option 2: Python Detector + MenuBar App (Recommended)

For a visual indicator in your macOS menubar:

1. Build and start the menubar app:

   cd HuddleIndicator
   swift build -c release
   ./.build/release/HuddleIndicator &

2. Start the Python detector:

   sudo python3 slack-huddle-detector-optimized.py

The detector will:

1. Request sudo access (needed for audio system monitoring)
2. Calibrate baseline (takes 6 seconds)
3. Begin monitoring for huddles
4. Display real-time status with indicators
5. Update menubar icon automatically

Output Example

🎧 Slack Huddle Detector - Optimized
==================================================
📊 Calibrating baseline (NOT in huddle)...
✅ Baseline score: 10

Monitoring for huddles...
  Start threshold: 35
  End: 70% drop from peak OR return to baseline+10

💤 No huddle | Score: 10→ | Monitoring... | Base:10 | 17:15:23

🟢 HUDDLE STARTED - 17:15:45
   Score: 65 (baseline: 10)
   • Audio power: 1
   • Slack assertions: 3
   • Audio units: 1

🎙️  IN HUDDLE | Score: 65→ | PWR:3 | AU:1 | Peak:65 | 17:15:48

🔴 HUDDLE ENDED - 17:18:32
   Score: 42 (peak was 65)
   New baseline: 42

How It Works

The detector monitors several macOS system indicators:

Primary Indicators

• Power Assertions: Slack creates audio power assertions during huddles
• Audio Units: AudioToolbox units are instantiated for audio processing
• HAL Plugins: Hardware Abstraction Layer plugins for audio I/O

Secondary Indicators

• Audio File Descriptors: Open file handles to audio devices
• IORegistry Clients: Audio engine client connections
• CoreAudio Connections: Direct connections to CoreAudio daemon

Detection Logic

Huddle Start Detection:

• Score exceeds baseline + 25 points (minimum 50)
• Requires 2 consecutive detections for confirmation

Huddle End Detection (any of):

• Score drops to 70% of peak huddle score
• Score returns within 10 points of baseline
• Strong downward trend detected

Scoring System

Indicator	Points	Description
Audio Power Assertions	25	System-level audio power management
Slack Assertions	20-60	Slack-specific power assertions (capped)
Audio Units	15-30	Audio processing units (capped)
HAL Plugins	10	Hardware audio plugins
Audio FDs	5	File descriptors above baseline
IO Clients	10	IORegistry audio clients
CoreAudio	10	CoreAudio daemon connections

Status Indicators

Command Line Display

• 🎙️ IN HUDDLE: Currently in a Slack huddle
• 💤 No huddle: Not in a huddle
• ↑↓→ Trend arrows: Score movement direction
• PWR: Power assertion count
• AU: Audio unit count
• FD: File descriptor count
• Peak: Highest score during current huddle
• Base: Current baseline score

MenuBar App

• ⚪️ (white circle): Not in huddle - visible on dark menubar
• 🎙️ (microphone): Currently in a Slack huddle
• ❌ (red X): Python detector not running
• Click icon: Shows detailed popup with metrics and connection status

Troubleshooting

"Sudo access required"

The detector needs sudo to access audio system information. This is safe and only reads system state.

False Positives

• Ensure you're not in a huddle during calibration
• The detector adapts its baseline after each huddle
• May trigger briefly when testing audio in Slack settings

Huddle End Not Detected

• The detector uses multiple methods to detect end
• If issues persist, try restarting the detector
• Score should drop significantly when huddle ends

High Baseline Score

• Normal if you have other audio applications running
• The detector uses relative changes, not absolute values
• Baseline updates after each huddle for adaptation

Technical Details

The detector uses macOS system frameworks to monitor audio state:

• IOKit/IORegistry for audio device state
• Power Management for assertions
• CoreAudio for audio session state
• File descriptors for audio device access

No network traffic monitoring or process memory inspection is performed.

Repository Structure

• slack-huddle-detector-optimized.py - Main Python detector (recommended)
• HuddleIndicator/ - Swift menubar app for visual status indicator
• failed attempts/ - Various experimental approaches and debug tools
  • slack-huddle-analyzer.py - Network connection analyzer
  • slack-huddle-iokit-fixed.py - IOKit-based detector (alternative)
  • slack-huddle-simple.py - Simple network-based detector
  • Various other debug and analysis tools

License

MIT License - See LICENSE file for details

Contributing

Contributions welcome! Please test changes thoroughly on macOS with Slack desktop app.

Acknowledgments

Developed through extensive testing and analysis of Slack's behavior on macOS during huddles.