azure-ai-anomalydetector-java

Build anomaly detection applications with Azure AI Anomaly Detector SDK for Java. Use when implementing univariate/multivariate anomaly detection, time-series analysis, or AI-powered monitoring.

Install with Tessl CLI

npx tessl i github:boisenoise/skills-collections --skill azure-ai-anomalydetector-java

What are skills?

2.33x

Quality

84%

Does it follow best practices?

Impact

98%

2.33x

Average score across 3 eval scenarios

SKILL.md

Review

Evals

Evaluation results

100%

56%

E-Commerce Traffic Anomaly Analyzer

Univariate batch anomaly detection

Criteria

Without context

With context

Correct Maven artifact

100%

Endpoint env var

100%

API key env var

100%

AnomalyDetectorClientBuilder usage

100%

UnivariateClient built

100%

AzureKeyCredential used

100%

TimeSeriesPoint construction

100%

Minimum 12 data points

100%

Granularity set

100%

Sensitivity set

100%

Batch detection method

50%

100%

HttpResponseException handling

100%

Without context: $0.7220 · 6m 55s · 30 turns · 38 in / 9,545 out tokens

With context: $0.4523 · 3m 49s · 20 turns · 182 in / 6,172 out tokens

100%

70%

Server Health Monitor with Trend Analysis

Streaming detection and change point detection

Criteria

Without context

With context

Correct Maven artifact

100%

Env var credentials

100%

UnivariateClient construction

100%

Last-point detection method

100%

isAnomaly check

50%

100%

Expected/margin values read

75%

100%

Change point method

100%

Change point options construction

100%

Confidence score accessed

50%

100%

Granularity alignment

100%

Sensitivity in range

100%

HttpResponseException handling

100%

Without context: $0.3150 · 3m 57s · 18 turns · 25 in / 4,927 out tokens

With context: $0.4285 · 4m 16s · 23 turns · 26 in / 5,307 out tokens

96%

42%

Manufacturing Equipment Anomaly Detection System

Multivariate model lifecycle management

Criteria

Without context

With context

Correct Maven artifact

100%

MultivariateClient built

100%

Env var credentials

100%

ModelInfo data source

100%

ModelInfo time range

100%

Sliding window in range

100%

trainMultivariateModel called

100%

Training status check

100%

TopContributorCount set

100%

Batch result polling

100%

Severity read

100%

Contributor interpretation

50%

100%

List models

100%

HttpResponseException handling

Without context: $0.4178 · 4m 2s · 18 turns · 25 in / 7,859 out tokens

With context: $0.4901 · 4m 17s · 21 turns · 294 in / 6,937 out tokens

Evaluated: 4 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

E-Commerce Traffic Anomaly Analyzer Server Health Monitor with Trend Analysis Manufacturing Equipment Anomaly Detection System

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.