Skip to content
Kure Monitor

Introduction

Kure Monitor is a Kubernetes health monitoring tool that helps you understand why your workloads fail. When a pod crashes, gets stuck pending, or can’t pull an image, Kure detects it instantly and provides AI-generated troubleshooting guidance to help you fix it fast. It also continuously scans your cluster for security misconfigurations and gives you a real-time overview of cluster topology — all from a single dashboard.

What’s in the box

Section titled “What’s in the box”
CapabilityDescription
AI-Powered TroubleshootingContextual fixes from pod events, logs, and manifest analysis using OpenAI, Anthropic, Groq, Google Gemini, GitHub Copilot, or Ollama
Real-time Failure DetectionCrashLoopBackOff, ImagePullBackOff, Pending, OOMKilled, FailedScheduling, FailedMount
Security Scanning50+ checks: privileged containers, host namespaces, dangerous capabilities, missing seccomp/AppArmor, root containers, RBAC misconfigs, untrusted registries, missing limits
Topology DiagramInteractive Kubernetes graph (per-namespace and per-workload), click nodes for manifests, click edges to focus paths
Mirror Pod TestingDeploy a temporary copy of a failing pod with the AI-suggested fix applied to verify before committing
Live Pod LogsStream logs in real-time with container selection
Export FindingsSecurity findings to CSV, JSON, and PDF
NotificationsSlack and Microsoft Teams webhooks
AuthUser accounts with read/write/admin roles, HttpOnly session cookies, rate-limited login
Prometheus Metrics/metrics endpoint with optional ServiceMonitor support

Limitations

Section titled “Limitations”

Kure is focused on failure diagnosis, not general observability:

  • No Prometheus dependency — Kure works standalone
  • Not a metrics platform — no time-series, no alerting rules, no historical dashboards
  • Not a log aggregator — logs are fetched on-demand, not stored or indexed
  • Single cluster only — monitors one Kubernetes cluster per installation

Kure complements your existing observability stack — it doesn’t replace it.

Architecture overview

Section titled “Architecture overview”
                         Kubernetes Cluster
  ┌────────────────────────────────────────────────────────────────┐
  │                                                                │
  │   ┌──────────────┐     ┌──────────────────┐     ┌───────────┐  │
  │   │    Agent     │────>│                  │<────│ Frontend  │  │
  │   │ (DaemonSet)  │ HTTP│     Backend      │ WS  │  (React)  │  │
  │   └──────┬───────┘     │    (FastAPI)     │     └───────────┘  │
  │          │             │                  │                    │
  │   ┌──────┴───────┐     │                  │                    │
  │   │   Security   │────>│                  │                    │
  │   │   Scanner    │ HTTP└────────┬─────────┘                    │
  │   └──────┬───────┘              │                              │
  │          │                      │                              │
  │          │              ┌───────┴──────┐                       │
  │    K8s API Server       │  PostgreSQL  │                       │
  │   (watch pods,          │   Database   │                       │
  │    events, nodes)       └──────────────┘                       │
  │                                                                │
  └────────────────────────────────────────────────────────────────┘
                                   │
                                   ▼
                          ┌──────────────────┐
                          │   LLM Provider   │
                          │ OpenAI/Anthropic │
                          │ Groq/Gemini/     │
                          │ Copilot/Ollama   │
                          └──────────────────┘

See Architecture for the full breakdown.

Next steps

Section titled “Next steps”