Habilidades azure-ai-anomalydetector-java
📦

azure-ai-anomalydetector-java

Seguro

Build Anomaly Detection Apps with Azure AI SDK for Java

Detect anomalies in time-series data using Azure Cognitive Services. This skill provides Java SDK patterns for univariate and multivariate anomaly detection with real-time streaming support.

Soporta: Claude Codex Code(CC)
🥉 75 Bronce
1

Descargar el ZIP de la skill

2

Subir en Claude

Ve a Configuración → Capacidades → Skills → Subir skill

3

Activa y empieza a usar

Pruébalo

Usando "azure-ai-anomalydetector-java". Detect anomalies in 30 days of daily website traffic data

Resultado esperado:

Anomaly detection completed. Found 3 anomalies: Day 7 (traffic spike: 15,200 vs expected 8,500), Day 18 (traffic drop: 2,100 vs expected 7,800), Day 25 (traffic spike: 18,900 vs expected 9,200). Confidence scores: 0.92, 0.87, 0.95.

Usando "azure-ai-anomalydetector-java". Monitor server CPU, memory, and disk I/O for anomalies

Resultado esperado:

Multivariate anomaly detected at 2024-01-15T14:32:00Z. Severity: 0.89. Top contributors: CPU usage (contribution: 0.45), Disk I/O wait (contribution: 0.32). Recommended action: Investigate potential resource contention.

Auditoría de seguridad

Seguro
v1 • 2/24/2026

All 35 static analysis findings are false positives. The external_commands detections misidentified Java code examples in markdown as Ruby shell execution. The network URLs are documentation placeholders for Azure blob storage. Environment variable access uses standard System.getenv() patterns recommended by Microsoft Azure SDK. The skill is documentation for Microsoft's official Azure AI Anomaly Detector SDK with no malicious intent.

1
Archivos escaneados
262
Líneas analizadas
0
hallazgos
1
Auditorías totales
No se encontraron problemas de seguridad
Auditado por: claude

Puntuación de calidad

38
Arquitectura
100
Mantenibilidad
87
Contenido
50
Comunidad
100
Seguridad
100
Cumplimiento de la especificación

Lo que puedes crear

IT Operations Monitoring

Detect anomalies in server metrics like CPU usage, memory consumption, and network traffic to identify potential outages before they impact users.

Financial Fraud Detection

Analyze transaction patterns across multiple variables to identify suspicious activities and potential fraud in real-time payment systems.

IoT Sensor Analytics

Monitor industrial equipment sensor data to detect early warning signs of equipment failure and enable predictive maintenance schedules.

Prueba estos prompts

Create Basic Univariate Client 
Create a Java class that initializes the Azure AI Anomaly Detector univariate client using environment variables for the endpoint and API key. Include error handling for missing credentials.
Implement Batch Anomaly Detection 
Write a method that performs batch anomaly detection on a time-series dataset. The method should accept a list of timestamps and values, configure detection sensitivity, and return a list of detected anomalies with their expected values and confidence scores.
Build Streaming Anomaly Detector 
Create a streaming anomaly detection service that processes incoming data points in real-time. Use the detectUnivariateLastPoint method to check each new data point and trigger alerts when anomalies exceed a severity threshold.
Train and Deploy Multivariate Model 
Implement a complete multivariate anomaly detection workflow: (1) prepare training data in Azure Blob Storage, (2) train a model with configurable sliding window, (3) poll for training completion, (4) run batch inference on new data, and (5) extract top contributing variables for each detected anomaly.

Mejores prácticas

  • Use at least 12 data points for univariate detection and align TimeGranularity with your actual data frequency for accurate results
  • Configure sensitivity between 80-95 for production usecases to balance false positives and missed anomalies
  • Handle HttpResponseException to gracefully manage API rate limits and service errors

Evitar

  • Do not use sensitivity values below 50 which may miss significant anomalies or above 99 which generates excessive false positives
  • Avoid calling detectUnivariateLastPoint without maintaining sufficient historical context for the algorithm
  • Do not hardcode API keys or endpoints in source code - always use environment variables or Azure Key Vault

Preguntas frecuentes

What is the minimum data requirement for anomaly detection?
Univariate detection requires at least 12 data points. Multivariate models perform best with 200-1000 training points depending on pattern complexity.
How do I choose between univariate and multivariate detection?
Use univariate for single metric analysis like CPU usage. Use multivariate when monitoring 2+ correlated signals where anomalies may only appear in their relationships.
Can I use this skill without an Azure subscription?
No. The Azure AI Anomaly Detector SDK requires an active Azure Cognitive Services subscription with the Anomaly Detector resource provisioned.
How long does multivariate model training take?
Training is asynchronous and typically completes within minutes depending on data volume. Use the polling pattern to check training status until the model reaches Ready state.
What time granularities are supported?
Supported granularities include: Yearly, Monthly, Weekly, Daily, Hourly, Minutely, Secondly, and MicroSecondly. Match granularity to your data collection frequency.
How do I handle authentication securely?
Use DefaultAzureCredential in production for managed identity support. For development, store credentials in environment variables or Azure Key Vault, never in source code.

Detalles del desarrollador

Autor

sickn33

Licencia

MIT

Repositorio

https://github.com/sickn33/antigravity-awesome-skills/tree/main/skills/azure-ai-anomalydetector-java

Ref.

main

Estructura de archivos

📄 SKILL.md