Vmanomaly

Operate VictoriaMetrics Anomaly Detection (vmanomaly) through its HTTP API. Use when checking health, versions, persisted-state compatibility, self-monitoring metrics, available models, model schemas, server queries, time-series characteristics, shared autotune, or asynchronous detection tasks. Also use for generating and validating vmanomaly configuration or alert-rule YAML. Triggers on vmanomaly API, anomaly tasks, model validation, model schema, compatibility, Temporal Envelope, time-series profiling, and vmanomaly autotune.

Design, tune, validate, and test VictoriaMetrics vmanomaly configurations for known metrics or LogsQL queries. Use when choosing static alerting versus ML, selecting a vmanomaly model, configuring Temporal Envelope, building deployment YAML, tuning anomaly sensitivity, or creating VMAlert rules. Trigger on anomaly configuration, model selection, shared autotune, seasonal anomaly detection, forecasting, or "how should I monitor this metric?" Do not use for open-ended discovery across unknown signals.

Audit an existing VictoriaMetrics vmanomaly configuration against runtime capabilities and real data. Use when reviewing model-data fit, scheduler cadence, persisted-state compatibility, cold start, excessive detections, missing anomalies, output cardinality, or upgrade readiness. Trigger on vmanomaly config review, false-positive investigation, detection quality, model effectiveness, compatibility, and "is my anomaly detection working?" Use vmanomaly-config instead when building a new configuration from scratch.