Lumino MCP Server

@spre-sre

リポジトリ

基本情報

カテゴリ

開発者ツール

トランスポート

stdio

作者

spre-sre

投稿者

geored

Lumino MCP Server について

AI/ML-powered diagnostic engine for SRE Observability on Konflux and OpenShift. It uses the Model Context Protocol (MCP) and 40+ tools to analyze logs, metrics, and traces, enabling automated RCA and predictive analysis.

#mcp #model-context-protocol #kubernetes #openshift#tekton#sre #observability #ai #machine-learning #devops#cicd#prometheus #log-analysis#root-cause-analysis#konflux#claude #anthropic #mcp-server

Lumino 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": {
    "lumino": {
      "type": "stdio",
      "command": "<ABSOLUTE_PATH_TO_LUMINO>/.venv/bin/python",
      "args": [
        "<ABSOLUTE_PATH_TO_LUMINO>/main.py"
      ],
      "env": {
        "PYTHONUNBUFFERED": "1"
      }
    }
  }
}

Lumino 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。

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is Lumino?

Lumino is an open source MCP (Model Context Protocol) server that empowers Site Reliability Engineers (SREs) and DevOps teams with intelligent observability, predictive analytics, and AI-driven automation across Kubernetes, OpenShift, and Tekton environments. It exposes 37 specialized tools through MCP for monitoring, analysis, troubleshooting, prediction, and simulation.

How to use Lumino?

Clone the repository, install Python dependencies with uv sync (or pip install -e .), and run python main.py. For Claude Code CLI users, configure a .mcp.json file pointing to the installed server. Alternatively, use mcpm install @spre-sre/lumino-mcp-server. After installation, authenticate to a Kubernetes/OpenShift cluster and approve the MCP server to start using Lumino tools.

Key features of Lumino

Kubernetes & OpenShift operations (namespace, pod, resource management)
Tekton pipeline intelligence (monitoring, root cause analysis, baselining)
Advanced log analysis (summarization, streaming, semantic search, anomaly detection)
Predictive & proactive monitoring (statistical anomaly detection, resource forecasting)
Event intelligence (smart retrieval, ML pattern detection, log-event correlation)
Simulation & what-if analysis (Monte Carlo, impact analysis, risk assessment)

Use cases of Lumino

Generate comprehensive RCA reports for failed pipeline runs
Predict resource bottlenecks across namespaces for the next 48 hours
Simulate configuration changes before deploying to production
Map live topology of services, deployments, and dependencies
Perform advanced forensic investigation with multi-faceted log/event analysis
Establish CI/CD pipeline performance baselines and detect deviations

よくある質問

What are the prerequisites for using Lumino?

Python 3.10+, an MCP client (Claude Desktop, Claude Code CLI, Gemini CLI, or Cursor IDE), and a valid Kubernetes/OpenShift kubeconfig with read permissions.

How does authentication work?

Lumino automatically detects Kubernetes configuration: in-cluster config when running inside a pod, or local kubeconfig (`~/.kube/config`) when running locally.

What transports are supported?

Local mode uses stdio transport for direct MCP client integration. When the `KUBERNETES_NAMESPACE` environment variable is set, the server switches to HTTP streaming transport for Kubernetes‑native operation.

What dependencies are required?

Core: Python 3.10+, `uv` or `pip`. Optional but recommended: `uv` for faster dependency management, `MCPM` for easier installation, and Prometheus for advanced metrics and forecasting.