VMAnomaly

About

This tool implements a Model Context Protocol (MCP) server designed specifically for VictoriaMetrics' VMAnomaly application. It provides a seamless interface, allowing AI assistants like Claude to interact directly with VMAnomaly's REST API. This integration empowers AI assistants to perform tasks such as health monitoring, model management (including listing, validating, and configuring over 10 anomaly detection model types), configuration generation, and alert rule generation for anomaly scoring. Additionally, the server embeds up-to-date VMAnomaly documentation, enabling AI assistants to perform offline full-text searches with fuzzy matching for enhanced observability and insight.

Key Features

• Health Monitoring: Check vmanomaly server status and build information.
• 2 GitHub stars
• Documentation Search: Full-text search across embedded vmanomaly documentation with fuzzy matching, even without online access.
• Model Management: List, validate, and configure over 10 anomaly detection model types (e.g., zscore, prophet, mad, holtwinters, isolation_forest).
• Configuration Generation: Generate complete vmanomaly YAML configurations.
• Alert Rule Generation: Create VMAlert rules for anomaly score alerting.

Use Cases

• Automating the generation of vmanomaly configurations and VMAlert rules for anomaly detection.
• Enabling AI assistants to monitor the health and performance of vmanomaly instances.
• Allowing AI assistants to manage anomaly detection models, including validation and configuration.