KubeBolt — Instant Kubernetes Monitoring & Management

Features

Built for teams without dedicated SRE

Connect your kubeconfig. See everything. Get actionable insights. No expertise required.

Zero Configuration

No CRDs, no YAML manifests, no sidecars. Works with EKS, GKE, AKS, k3s, Docker Desktop out of the box.

Multi-Cluster

All kubeconfig contexts discovered automatically. Switch between dev, staging, and production in one click.

Cluster Topology Map

Interactive visualization with Grid and Flow layouts. Trace the path from Ingress to Pod. Namespace grouping and filtering.

Insights Engine

12 rules detect crash loops, OOM kills, CPU throttling, HPA saturation. Actionable recommendations, not raw metrics.

Authentication & Roles

Built-in login with Admin, Editor, and Viewer roles. User management UI. No OAuth setup required. JWT sessions with httpOnly cookies.

RBAC-Aware

Auto-detects permissions via SelfSubjectAccessReview. Adapts to cluster-admin, read-only, or namespace-scoped ServiceAccounts.

Real-Time Updates

WebSocket-powered via Kubernetes shared informers. Pod changes, scaling events, and new insights the moment they happen.

Cluster Actions

Pod terminal, file browser, port-forward, restart, scale, YAML edit, describe, delete — all from the browser.

Install

Choose your method

From a single command to full Kubernetes-native deployment.

StandardIn-Cluster

Helm Chart

Production deployment. Configurable RBAC, Ingress, auth, resources. OCI chart on GHCR.

helm install kubebolt \
  oci://ghcr.io/clm-cloud-solutions/kubebolt/helm/kubebolt

kubectl port-forward svc/kubebolt 3000:80

# With custom admin password
helm install kubebolt \
  oci://ghcr.io/clm-cloud-solutions/kubebolt/helm/kubebolt \
  --set auth.adminPassword=YourSecurePassword

EasyAny OS

Docker

Single container with embedded frontend, or full stack with Docker Compose. Multi-arch (amd64/arm64).

# Single container (recommended)
docker run -p 3000:3000 -v ~/.kube:/root/.kube:ro \
  ghcr.io/clm-cloud-solutions/kubebolt:latest

# Or full stack with Docker Compose
cd deploy && docker compose up -d

EasymacOSLinuxWindows

Single Binary

Download and run. API server and web UI in a single executable. Embedded frontend, one port. Auto-loads .env for configuration.

# macOS Apple Silicon
curl -LO https://github.com/clm-cloud-solutions/kubebolt/releases/latest/download/kubebolt-darwin-arm64
chmod +x kubebolt-darwin-arm64 && mv kubebolt-darwin-arm64 /usr/local/bin/kubebolt

# Run
kubebolt --kubeconfig ~/.kube/config

EasymacOSLinux

Homebrew

Install and update via Homebrew. Automatic version management.

brew install clm-cloud-solutions/tap/kubebolt
kubebolt --kubeconfig ~/.kube/config

AdvancedIn-Cluster

Kubernetes Operator

Declarative management via CRD. Handles upgrades, config, and self-healing.

kubectl apply -f https://get.kubebolt.dev/operator.yaml
kubectl apply -f kubebolt-instance.yaml

Coming soon — Phase 1.9

Standardkubectl

kubectl Plugin (krew)

Run KubeBolt as a kubectl subcommand. Uses the current kubectl context.

kubectl krew index add clm \
  https://github.com/clm-cloud-solutions/krew-index.git
kubectl krew install clm/kubebolt
kubectl kubebolt

Need help? Check the full documentation.

AI Copilot

Talk to your cluster in natural language

Press Cmd+J and ask anything. 17 native tools query your live cluster data. Contextual "Ask Copilot" buttons on insights, resources and events. Auto-compact keeps long sessions usable. Multi-provider, BYO key, RBAC-aware.

KubeBolt Copilot Claude Sonnet 4

Why is the api-gateway pod crashing in production?

get_pod_logs get_events get_resource_detail

Found 3 issues with production/api-gateway:

1. OOMKilled 2 min ago — memory limit 512Mi insufficient.
2. Readiness probe failing on /health after restart.
3. Last deploy changed DB_POOL_SIZE from 10 → 50.

Recommendation: increase memory to 1Gi and revert DB_POOL_SIZE.

Ask about your cluster... Cmd+J

17 Tools

Live cluster queries

BYO Key

Claude · GPT · Ollama · vLLM

Auto-compact

Long sessions stay cheap

Ask Copilot

Contextual from insights & resources

Privacy-First

Secrets redacted · RBAC-aware

Admin

Know exactly what your Copilot costs

Every chat session persists locally. Track tokens, cache hit rate and estimated USD by provider/model. 30-day retention in BoltDB. Zero telemetry, zero SaaS — the data stays with you.

Usage analytics

Sessions, tokens billed, cache hit rate, avg duration, compact events — drill down per session.

Estimated cost (USD)

Built-in pricing table for Anthropic and OpenAI. Approximate since BYOK, accurate enough to spot runaway spend.

Tool breakdown

Calls, tokens, error rates per tool. Catch tools that return too much or fail too often.

Scissors & compact

Manual "new session with summary" one click away, plus automatic compaction at 80% of the context budget.

Notifications

Alerts where your team already is

Slack, Discord and email channels for insight alerts. Global master toggle for maintenance windows, base URL for deep links back to KubeBolt, resolved-insight alerts, and email digest modes (instant / hourly / daily).

Slack · Discord

Rich webhook messages with severity colour-coding and deep-link buttons back to the resource.

Email with digest

SMTP delivery with three modes: instant for criticals, hourly or daily digests to keep noise low.

Global controls

Master enable, min severity threshold, dedup cooldown, resolved-insight notifications. All from env vars or the admin UI.

See your entire Kubernetes cluster. Instantly.