MCP.so
ログイン

VictoriaMetrics MCP Server

@yincongcyincong

基本情報

カテゴリ

その他

トランスポート

stdio

作者

yincongcyincong

VictoriaMetrics MCP Server について

MCP Server for the VictoriaMetrics.

#metrics#mcp#victoriametrics#mcp-server

VictoriaMetrics MCP Server の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "VictoriaMetrics-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "@smithery/cli",
        "install",
        "@yincongcyincong/victoriametrics-mcp-server",
        "--client",
        "claude"
      ]
    }
  }
}

VictoriaMetrics MCP Server のツール

ツール一覧はまだありません

この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。

上のボタンで一時的なサンドボックスでこのサーバーを実行し、ツール一覧を取得できます。

概要

What is VictoriaMetrics MCP Server?

An MCP server that provides direct read and write access to VictoriaMetrics databases, enabling large language model agents to query, write, and manage time-series data via PromQL and native APIs. It is designed for developers and data analysts integrating AI assistants with monitoring or observability stacks.

How to use VictoriaMetrics MCP Server?

Install via Smithery (npx -y @smithery/cli install ... --client claude) or configure directly in Claude Desktop by adding the NPX command with the required VM_URL, VM_SELECT_URL, and VM_INSERT_URL environment variables. Invoke tools such as vm_data_write, vm_prometheus_write, vm_query_range, vm_query, vm_labels, and vm_label_values.

Key features of VictoriaMetrics MCP Server

  • Write time-series data to VictoriaMetrics via the vm_data_write tool.
  • Import Prometheus exposition format data with vm_prometheus_write.
  • Query metric values over a time range using PromQL (vm_query_range).
  • Get current metric values with vm_query.
  • Retrieve unique label names and label values (vm_labels, vm_label_values).
  • Optional resolution step and evaluation timestamps for queries.

Use cases of VictoriaMetrics MCP Server

  • AI-powered dashboard that writes custom metrics to VictoriaMetrics in real time.
  • Automated incident response that queries recent metric spikes via PromQL.
  • Data exploration assistant that retrieves and analyzes time-series labels.
  • Observability chatbot that imports Prometheus export data for historical analysis.

よくある質問

1

What data can I write and query with this server?

You can write metrics with tags, values, and timestamps, or import Prometheus exposition format strings. Query with PromQL expressions over time ranges or at specific timestamps.

2

What environment variables are required?

The server needs `VM_URL`, and optionally `VM_SELECT_URL` and `VM_INSERT_URL`. All must be set when running the NPX command.

3

How do I install the server?

Use the Smithery CLI for Claude Desktop or add the NPX configuration block to your MCP configuration file.

4

Does the server support authentication or custom transports?

The README does not mention authentication or custom transports; only a basic HTTP-based setup with environment variables is documented.

5

Can I debug the server locally?

Yes, using the MCP inspector: `npx @modelcontextprotocol/inspector -e VM_URL=http://127.0.0.1:8428 node src/index.js`.

コメント

「その他」の他のコンテンツ