VictoriaMetrics MCP Server

Name: VictoriaMetrics MCP Server
Author: VictoriaMetrics-Community

@VictoriaMetrics-Community

Repository Homepage

About VictoriaMetrics MCP Server

The implementation of Model Context Protocol (MCP) server for VictoriaMetrics

#victoriametrics #mcp-server #monitoring #observability #mcp #ai #metrics

Basic information

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "victoriametrics": {
      "command": "/path/to/mcp-victoriametrics",
      "env": {
        "VM_INSTANCE_ENTRYPOINT": "<YOUR_VM_INSTANCE>",
        "VM_INSTANCE_TYPE": "<YOUR_VM_INSTANCE_TYPE>",
        "VM_INSTANCE_BEARER_TOKEN": "<YOUR_VM_BEARER_TOKEN>"
      }
    }
  }
}

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is VictoriaMetrics MCP Server?

The VictoriaMetrics MCP Server is an implementation of the Model Context Protocol (MCP) that provides an interface to VictoriaMetrics instances. It enables AI assistants and tools to query metrics, explore data, search documentation, and perform monitoring and debugging tasks using VictoriaMetrics read-only APIs.

How to use VictoriaMetrics MCP Server?

Install via pre-built binaries, Docker, or build from source. Configure environment variables such as VM_INSTANCE_ENTRYPOINT and VM_INSTANCE_TYPE to point to your VictoriaMetrics instance. Then connect an MCP-compatible client to the server (stdio, SSE, or Streamable HTTP modes). A public test instance is available at https://play-mcp.victoriametrics.com/mcp.

Key features of VictoriaMetrics MCP Server

Query metrics and explore data, including graph rendering
List, export metrics, labels, label values, and series
Analyze and test alerting and recording rules
Show VictoriaMetrics instance configuration parameters
Explore cardinality and metrics usage statistics
Debug relabeling, downsampling, and retention policies
Embedded up-to-date documentation searchable offline

Use cases of VictoriaMetrics MCP Server

AI-assisted monitoring and observability of VictoriaMetrics instances
Automated debugging and investigation of metric data
Integration with MCP-compatible clients for natural language queries
Testing and validating alerting and recording rules without a UI

FAQ from VictoriaMetrics MCP Server

What are the prerequisites to run the server?

You need a VictoriaMetrics instance (single-node or cluster) or a VictoriaMetrics Cloud account. To build from source, Go 1.26+ is required; pre-built binaries do not require Go.

What transport modes are supported?

Three modes: stdio (default for CLI integration), sse (Server-Sent Events over HTTP), and http (Streamable HTTP). Configure via the MCP_SERVER_MODE environment variable.

How can I authenticate to my VictoriaMetrics instance?

Use the VM_INSTANCE_BEARER_TOKEN environment variable for a bearer token, or provide custom HTTP headers via VM_INSTANCE_HEADERS. For VictoriaMetrics Cloud, use VMC_API_KEY.

Can I use it with VictoriaMetrics Cloud?

Yes. Set the VMC_API_KEY environment variable with an API key from the VictoriaMetrics Cloud Console. No VM_INSTANCE_ENTRYPOINT is needed when using VMC_API_KEY.