VMA-nomaly FAQs

Name: VMA-nomaly
Author: VictoriaMetrics

Question 1

Can VMA-nomaly help with documentation and learning about `vmanomaly`?

Accepted Answer

Yes, VMA-nomaly includes embedded, up-to-date `vmanomaly` documentation. It offers a full-text search capability with fuzzy matching, allowing users to find relevant information quickly without requiring an online connection.

Question 2

What is VMA-nomaly and its primary purpose?

Accepted Answer

VMA-nomaly is a Model Context Protocol (MCP) server that provides seamless integration between AI assistants and VictoriaMetrics Anomaly Detection (`vmanomaly`). Its primary purpose is to enable AI assistants to interact with `vmanomaly`'s REST API for enhanced anomaly detection, model management, and observability.

Question 3

What are the core requirements to use VMA-nomaly?

Accepted Answer

To use VMA-nomaly, you need an existing `vmanomaly` instance (version 1.28.3 or higher) with REST API access. If you're building VMA-nomaly from source, Go 1.24 or higher is also required.

Question 4

How does VMA-nomaly improve anomaly detection workflows?

Accepted Answer

VMA-nomaly streamlines anomaly detection by allowing AI assistants to perform key tasks such as checking server health, listing and configuring anomaly detection models, generating `vmanomaly` YAML configurations, and creating `vmalert` alerting rules based on anomaly scores. This automates and simplifies complex operations.

Question 5

What kind of anomaly detection models can be managed through VMA-nomaly?

Accepted Answer

VMA-nomaly supports managing various anomaly detection models available in `vmanomaly`, including popular ones like `zscore_online` and `prophet`. AI assistants can list, validate, and configure these models via the MCP server.

VMA-nomaly

VMA-nomaly

Key Features

Use Cases

Key Features

Use Cases