Agent skill

scrypted

Build Scrypted plugins for home automation. Create camera integrations, automations, and smart home bridges. Use for HomeKit, Google Home, and Alexa integrations.

View SKILL.md on GitHub Repository
Stars 163
Forks 31

Install this agent skill to your Project

npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/data/scrypted

SKILL.md

Scrypted Skill

Complete guide for Scrypted - the home video integration platform.

Quick Reference

Key Features

Feature Description
HKSV HomeKit Secure Video with local processing
Google Home Camera streaming to Google ecosystem
Alexa Camera streaming to Amazon ecosystem
NVR Built-in recording and playback
AI Detection Object detection (person, car, animal)

Access Points

http://<scrypted-ip>:10443  # Web interface
https://<scrypted-ip>:10443 # HTTPS

1. Installation

Docker Installation (Recommended)

yaml
services:
  scrypted:
    image: koush/scrypted:latest
    container_name: scrypted
    restart: unless-stopped
    network_mode: host
    volumes:
      - ./volume:/server/volume
      - /var/run/dbus:/var/run/dbus
      - /var/run/avahi-daemon/socket:/var/run/avahi-daemon/socket
    devices:
      # For hardware transcoding
      - /dev/dri:/dev/dri
      # For Coral TPU
      # - /dev/bus/usb:/dev/bus/usb
    environment:
      - SCRYPTED_WEBHOOK_UPDATE_AUTHORIZATION=Bearer YOUR_TOKEN
      - SCRYPTED_WEBHOOK_UPDATE=http://localhost:10444/v1/update

Docker Compose with GPU

yaml
services:
  scrypted:
    image: koush/scrypted:latest
    container_name: scrypted
    restart: unless-stopped
    network_mode: host
    volumes:
      - ./volume:/server/volume
    devices:
      - /dev/dri:/dev/dri  # Intel/AMD GPU
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: 1
              capabilities: [gpu]

Raspberry Pi

bash
# Install via script
curl -s https://raw.githubusercontent.com/koush/scrypted/main/install/docker/install-scrypted-docker-compose.sh | bash

npm Installation

bash
# Install Node.js 18+
npm install -g @scrypted/server

# Run
npx scrypted serve

2. Initial Setup

First Login

  1. Navigate to https://<ip>:10443
  2. Accept self-signed certificate warning
  3. Create admin account
  4. Complete setup wizard

Add Camera

  1. Click "+" to add device
  2. Select camera brand or "Generic RTSP"
  3. Enter camera URL and credentials
  4. Configure detection and recording options

Generic RTSP Camera

# RTSP URL format
rtsp://username:password@192.168.1.100:554/stream1

# Common paths:
# Hikvision: /Streaming/Channels/101
# Dahua: /cam/realmonitor?channel=1&subtype=0
# Amcrest: /cam/realmonitor?channel=1&subtype=0
# Reolink: /h264Preview_01_main

3. HomeKit Secure Video (HKSV)

Enable HomeKit Plugin

  1. Plugins > HomeKit
  2. Install "HomeKit" plugin
  3. Scan QR code with Apple Home app

Configure HKSV

  1. Select camera in Scrypted
  2. Enable "HomeKit Secure Video" extension
  3. In Apple Home: Camera settings > Recording Options
  4. Enable "Stream & Allow Recording"

HKSV Requirements

  • Apple Home Hub (HomePod/Apple TV)
  • iCloud+ subscription (50GB minimum)
  • Supported resolution (1080p max for HKSV)

Optimize for HKSV

yaml
# Camera settings in Scrypted
Video Codec: H.264
Resolution: 1920x1080 (or lower)
Framerate: 30 fps
Bitrate: 2000-4000 kbps
Audio: AAC (if supported)

4. Google Home Integration

Install Google Home Plugin

  1. Plugins > Install "Google Home"
  2. Link account in Google Home app
  3. Add Scrypted as linked service

Configure Cameras

  1. Enable "Google Home" extension on camera
  2. Camera appears in Google Home app
  3. View on Nest displays and TV

Supported Features

  • Live streaming
  • Talk back (2-way audio)
  • Event notifications
  • Doorbell announcements

5. Alexa Integration

Install Alexa Plugin

  1. Plugins > Install "Alexa"
  2. Link in Alexa app
  3. Discover devices

Configure Cameras

  1. Enable "Alexa" extension on camera
  2. View on Echo Show devices
  3. Enable doorbell announcements

6. NVR (Recording)

Enable NVR Plugin

  1. Plugins > Install "NVR"
  2. Configure storage location
  3. Set retention policies

Storage Configuration

yaml
# In Scrypted NVR settings
Storage Path: /server/volume/recordings
Retention Days: 7
Pre-Event Buffer: 10 seconds
Post-Event Buffer: 10 seconds

Recording Modes

Mode Description
Continuous Record everything
Event Record on motion/detection
Smart Record on AI detection (person, car, etc.)

View Recordings

  1. Select camera > Timeline
  2. Use timeline scrubber
  3. Download clips as needed

7. Object Detection

Built-in Detection

  • Motion detection (basic)
  • Uses camera's built-in analytics when available

OpenCV Plugin

yaml
# Install OpenCV plugin for AI detection
Plugins > OpenCV

# Detects:
- Person
- Car/Vehicle
- Animal
- Face

Coral TPU Plugin

yaml
# For Google Coral TPU acceleration
Plugins > TensorFlow Lite

# Docker device mapping
devices:
  - /dev/bus/usb:/dev/bus/usb  # USB Coral
  - /dev/apex_0:/dev/apex_0     # PCIe Coral

OpenVINO Plugin (Intel)

yaml
# For Intel CPU/GPU acceleration
Plugins > OpenVINO

# Best for Intel NUC, 6th gen+ processors

Detection Zones

  1. Select camera > Detection
  2. Draw zones on camera view
  3. Configure objects to detect per zone
  4. Set sensitivity and thresholds

8. Two-Way Audio

Requirements

  • Camera with audio input
  • Speaker/microphone support
  • Compatible platform (HomeKit, Google, Alexa)

Configuration

  1. Enable audio on camera
  2. Configure audio codec (AAC preferred)
  3. Test in platform app

Doorbell Configuration

yaml
# Enable doorbell button detection
1. Camera settings > Doorbell
2. Map button press event
3. Configure chime behavior

9. Advanced Configuration

Multiple Streams

yaml
# Configure separate streams for different purposes
Main Stream: rtsp://....:554/stream1 (high quality, recording)
Sub Stream: rtsp://....:554/stream2 (low quality, detection)

# In camera settings
Recording Stream: Main
Detection Stream: Sub
Remote Stream: Sub

Rebroadcast Plugin

yaml
# Create reliable RTSP streams
Plugins > Rebroadcast

# Provides:
- RTSP re-streaming
- HLS output
- WebRTC output
- MPEG-DASH output

Script Plugin

javascript
// Custom automation script example
const camera = device;

camera.on('motionDetected', async () => {
  console.log('Motion detected!');
  // Trigger other actions
});

Webhooks

yaml
# Configure webhook notifications
1. Camera > Webhooks
2. Add webhook URL
3. Select events to trigger

# Webhook payload includes:
- Event type
- Camera ID
- Timestamp
- Thumbnail (optional)

10. Troubleshooting

Common Issues

Camera not connecting:

bash
# Test RTSP stream
ffprobe rtsp://user:pass@192.168.1.100:554/stream1

# Check Scrypted logs
docker logs scrypted

# Verify network connectivity
ping 192.168.1.100

Poor streaming quality:

yaml
# Optimize camera settings
- Reduce resolution to 1080p
- Lower bitrate (2-4 Mbps)
- Use H.264 (not H.265)
- Enable hardware transcoding

HomeKit issues:

bash
# Reset HomeKit pairing
1. Remove bridge from Apple Home
2. Scrypted > HomeKit plugin > Reset
3. Re-pair with QR code

# Check mDNS
avahi-browse -rt _hap._tcp

High CPU usage:

yaml
# Enable hardware transcoding
1. Verify GPU passthrough in Docker
2. Install appropriate plugin (Intel/NVIDIA/AMD)
3. Select hardware transcoder in camera settings

# Reduce detection resolution
- Use sub-stream for detection
- Lower detection FPS

Logs

bash
# Docker logs
docker logs -f scrypted

# Scrypted console
# Web UI > Console tab

# Enable debug logging
# Settings > Logging > Debug

11. Performance Optimization

Hardware Transcoding

yaml
# Intel Quick Sync
devices:
  - /dev/dri:/dev/dri

# NVIDIA
deploy:
  resources:
    reservations:
      devices:
        - driver: nvidia
          count: 1
          capabilities: [gpu]

Memory Management

yaml
# Docker memory limits
deploy:
  resources:
    limits:
      memory: 4G

Network Optimization

  1. Use wired connection for Scrypted server
  2. Place cameras on dedicated VLAN
  3. Use PoE for reliable camera power
  4. Configure QoS for video traffic

Best Practices

  1. Use sub-streams for detection (saves CPU)
  2. Enable hardware transcoding when available
  3. Use HKSV for Apple users (end-to-end encrypted)
  4. Set retention policies to manage storage
  5. Configure zones to reduce false positives
  6. Regular backups of Scrypted configuration
  7. Update plugins regularly for fixes and features
  8. Monitor resources - cameras can be CPU intensive
  9. Use Coral TPU for best AI detection performance
  10. Separate network for cameras (security/performance)

Didn't find tool you were looking for?

Be as detailed as possible for better results