Dashboards — Multi-Cluster Kubernetes Monitoring

KubeStellar Console has 28+ dashboards for multi-cluster Kubernetes monitoring. Each dashboard gives you fleet-wide visibility into a specific operational area across all your connected clusters.

Main Dashboard

Route: /

This is your home page. It shows:

Overview of all your clusters
Cards you’ve chosen to see
Quick stats at the top
AI suggestions for what to look at

The main dashboard learns what you care about and shows those things first.

Dedicated Dashboards (28)

Clusters Dashboard

Route: /clusters

See all your Kubernetes clusters:

Which clusters are healthy (green)
Which clusters have problems (red)
Which clusters are offline (gray)
Quick links to each cluster’s native console

Best for: Checking if all your clusters are working

Workloads Dashboard

Route: /workloads

See all your running applications:

Deployments and their status
Pods that are having problems
Which apps are healthy

Best for: Making sure your applications are running

Compute Dashboard

Route: /compute

See your compute resources:

How many CPUs you have
How much memory is available
GPU usage (important for AI workloads!)
Top pods using resources

Best for: Checking if you have enough resources

Storage Dashboard

Route: /storage

See your storage:

Persistent Volume Claims (PVCs)
Storage classes
Which volumes are bound or pending

Best for: Managing disk space for your apps

Network Dashboard

Route: /network

See your networking:

Services and their types
LoadBalancers
Ingresses
Endpoints

Best for: Understanding how traffic flows

Events Dashboard

Route: /events

See what’s happening:

Recent events from all clusters
Warnings that need attention
Normal events
Filter by time or type

Best for: Troubleshooting when something goes wrong

Security Dashboard

Route: /security

Find security issues:

Containers running as root
Privileged containers
Missing security contexts
Critical and high severity issues

Best for: Keeping your clusters secure

Security Posture Dashboard

Route: /security-posture

Comprehensive security scanning, vulnerability assessment, and policy enforcement:

Compliance Score: Overall security score across all clusters (e.g., 78%)
Total Checks: Count of all security checks performed (405+)
Benchmark Scores: CIS, NSA, PCI-DSS benchmark compliance percentages
Policy Violations: Real-time violation tracking with severity breakdown
OPA Policies: Create, manage, and enforce OPA Gatekeeper policies with AI-assisted policy generation
Kyverno Policies: Install and manage Kyverno for Kubernetes-native policy management
Vulnerability Scanning: Critical and high CVE tracking across container images
Kubescape Integration: Automated security posture scanning with 80%+ benchmark scores

New in March 2026:

AI-driven Create Policy modal for natural language policy generation
Parallel cluster checks for faster policy evaluation across many clusters
Two-phase loading: policy metadata loads instantly, violations populate in background

Best for: Enterprise security compliance and policy enforcement

GitOps Dashboard

Route: /gitops

Manage GitOps:

Helm releases and their status (295 releases)
Kustomizations
ArgoCD applications with Sync Now button for immediate sync
GitOps Restart tab in ArgoCD drilldown for declarative application restarts
Drift detection with deployment status tracking (391 deployments)
Operator sync status (60 operators, 4 pending)

Best for: Managing deployments from git

Alerts Dashboard

Route: /alerts

Manage alerts:

Firing alerts with type-aware deduplication
Pending alerts
Alert rules you’ve created (4 enabled, 3 disabled)
Resolved alerts (119 resolved)
Falco integration for runtime security monitoring
Warning Events feed
macOS native notifications with click-to-navigate

Best for: Knowing when things need attention

GPU Reservations Dashboard

Route: /gpu-reservations

Schedule and manage GPU resources across your clusters with five dedicated tabs:

Overview: Total GPUs, availability, utilization donut chart, GPU types breakdown, allocation by cluster
Calendar: Visual calendar view of GPU reservations and availability windows
Reservations: Active and pending GPU reservations with details
Inventory: Full GPU inventory across all clusters with type, count, and status
Dashboard: Customizable card-based view of GPU metrics

Key features:

Create GPU reservations with namespace, cluster, and time range
View GPU usage by namespace with donut chart breakdowns
Track 12+ GPU types: NVIDIA A100/H100/A10G/V100/T4, Google TPU v4/v5e, Intel Gaudi2/AIU/Data Center GPU Max/Flex, IBM AIU
GPU Allocation by Cluster bar chart for capacity planning

Best for: AI/ML teams sharing GPUs across multi-cloud environments

Cost Management Dashboard

Route: /cost

Track your spending:

Total estimated cost
Cost per cluster
Cost by resource type (CPU, memory, storage)
OpenCost and Kubecost integration

Best for: Controlling cloud spending

Compliance Dashboard

Route: /compliance

Comprehensive security scanning, vulnerability assessment, and policy enforcement across your entire fleet:

Compliance Score: Composite score computed from OPA Gatekeeper, Kyverno, Kubescape, and Trivy data
Stats Overview: Score, total checks, passing, failing, CIS/NSA/PCI-DSS benchmarks, Gatekeeper violations, Kyverno violations, Kubescape score
Policy Violations: Aggregated violations from OPA + Kyverno with per-policy cluster attribution
Fleet Compliance Heatmap: Clusters × compliance tools grid with color-coded status (green/yellow/red). Shows install CTA icons when Kyverno/Kubescape/Trivy aren’t detected, linking to AI Mission install flows
Compliance Drift: Flags clusters deviating >1 standard deviation from fleet baseline compliance scores
Cross-Cluster Policy Comparison: Select up to 4 clusters and compare policy pass/fail in a table sorted by most discrepancies
Kyverno Policies: Live per-cluster policy data via CRD auto-detection
Kubescape Scan: Per-cluster framework scores via API aggregation or CRD check
Trivy Scan: Per-cluster vulnerability counts by severity
Cert Manager: Certificate expiry tracking across clusters

New in March 2026:

All compliance cards rewritten to use live per-cluster data (previously static demo data)
New useKyverno, useTrivy, useKubescape hooks with CRD auto-detection, localStorage caching, and demo fallback
3 new cross-cluster comparison cards (Fleet Compliance Heatmap, Compliance Drift, Cross-Cluster Policy Comparison)
Install CTA icons in heatmap headers link to AI Missions for one-click tool installation

Best for: Enterprise security compliance, fleet-wide posture assessment, and identifying outlier clusters

Logs Dashboard

Route: /logs

View logs:

Container logs from any pod
Filter by namespace or pod
Search log content

Best for: Debugging application issues

Helm Releases Dashboard

Route: /helm

Manage Helm:

All Helm releases
Release history
Values comparison
Available upgrades

Best for: Managing Helm deployments

Services Dashboard

Route: /services

See all services:

ClusterIP services
LoadBalancer services
NodePort services
Endpoints

Best for: Understanding service networking

Operators Dashboard

Route: /operators

Manage operators:

OLM operators
Subscriptions
Available updates

Best for: Managing cluster extensions

Nodes Dashboard

Route: /nodes

See your nodes:

Node health status
Resource usage per node
Node labels and taints

Best for: Infrastructure monitoring

Deployments Dashboard

Route: /deployments

Focus on deployments:

All deployments across clusters
Replica counts
Rollout status

Best for: Application deployment status

Pods Dashboard

Route: /pods

Focus on pods:

All pods across clusters
Pod status
Restart counts
Resource usage

Best for: Detailed pod troubleshooting

AI/ML Dashboard

Route: /ai-ml

Monitor AI and Machine Learning workloads:

llm-d inference stack monitoring (Request Flow, KV Cache, EPP Routing)
Prefill/Decode disaggregation metrics
llm-d benchmarks and comparisons
ML Jobs and Notebooks
GPU Overview with type breakdown
Hardware Health monitoring
Node Offline Detection with AI predictions

Best for: Managing AI/ML infrastructure and LLM serving stacks

llm-d Benchmarks Dashboard

Route: /llm-d-benchmarks

Performance tracking across clouds and accelerators for the llm-d inference stack:

Nightly E2E Status: Real-time pass rates across 16 guides on OCP, GKE, and CKS platforms with per-guide green/red dot matrix and AI-generated summary
Pareto Frontier: Tabbed chart views comparing throughput vs. latency tradeoffs across configurations
Leaderboard: Ranked model/configuration comparison with pagination
Benchmark Hero: Summary metrics from the latest benchmark runs
Live Data: Streams benchmark results from Google Drive via SSE (Server-Sent Events) with automatic fallback to demo data

The Nightly E2E Status card features:

89% overall pass rate with 16 active guides
Per-platform breakdown (OCP, GKE, CKS) with individual pass rates
Sparkline trend graph showing pass rate over time
AI summary with duration, model, and GPU information
Detail panel with per-guide status and last run timestamps

Best for: Tracking llm-d inference stack performance and CI health across platforms

AI Agents Dashboard

Route: /ai-agents

Manage Kagenti AI agents:

Agent fleet overview across clusters with on/off toggle and “Live” indicator
MCP Tool Registry with searchable tool listing
Agent Discovery with skill tags and cost analysis capabilities
Build Pipeline with recent build history and status
Framework breakdown (LangGraph, CrewAI, AG2)
Agent build status and history
SPIFFE identity coverage
Per-agent replica status and cluster placement
Agent memory persistence across sessions

Best for: Deploying, securing, and monitoring AI agents

CI/CD Dashboard

Route: /ci-cd

Monitor continuous integration and deployment:

PROW CI status and success rates
PROW Jobs with type/state filtering and pagination
PROW CI Monitor with success rate tracking
PROW revision history
Helm release tracking (295 releases)
Kustomize and ArgoCD sync status
Operator deployments (5,412 total, 287 deployed, 4 pending)

Best for: Monitoring CI/CD pipelines and PROW test infrastructure

Deploy Dashboard

Route: /deploy

Multi-cluster deployment management:

Workloads overview with drag-to-deploy (659 total, 553 unique)
Cluster Groups for targeting deployments
Deployment Missions with AI-assisted rollouts and Mission Browser
Mission Browser with Installer and Solution tabs for discovering pre-built missions
Deep-linking and sharing for missions with OAuth flow support
Saved Missions panel for quick access
Resource Marshall for workload placement
Deployment history and rollback

Best for: Deploying and managing workloads across multiple clusters

Data Compliance Dashboard

Route: /data-compliance

Monitor data compliance:

Data classification status
Compliance checks and violations
Policy enforcement across clusters

Best for: Meeting data governance requirements

Insights Dashboard

Route: /insights

Cross-cluster correlation and pattern detection using heuristic algorithms and optional AI enrichment:

Stats Overview: Clusters, insights detected, critical count, warnings count
7 Insight Cards:
- Cross-Cluster Event Correlation: Detects simultaneous warning events across clusters within a 5-minute window, suggesting common upstream causes
- Resource Imbalance Detector: Identifies CPU/memory imbalances across the fleet (e.g., one cluster at 87% while others sit at 22%)
- Config Drift Heatmap: Visualizes configuration differences between clusters for the same workloads
- Cluster Delta Detector: Tracks changes in cluster state over time, flagging unusual deltas
- Restart Correlation Matrix: Correlates pod restart patterns across clusters
- Cascade Impact Map: Shows how failures in one cluster propagate to dependent services
- Deployment Rollout Tracker: Monitors deployment rollout progress across clusters
AI Enrichment: When kc-agent is connected, heuristic insights are enriched with AI-generated root cause analysis, remediation suggestions, and confidence scores
Insight Source Badge: Each insight shows (H) for heuristic or (AI) for AI-enriched
Remediation Blocks: AI suggestions appear as blue-highlighted blocks with actionable remediation steps

New in March 2026:

AI enrichment via useInsightEnrichment hook with debounced requests, WebSocket broadcast, and TTL cache
Backend InsightWorker with rule-based fallback when no AI provider is connected
Remediation blocks added to all 7 insight cards
All cards respect global cluster filters

Best for: Identifying cross-cluster patterns that are invisible when monitoring clusters individually

Arcade Dashboard

Route: /arcade

Take a break with Kubernetes-themed games:

21 games including AI Checkers, Kube Chess, Container Tetris, Sudoku
High scores saved locally
Multiple themes available

Best for: Team building and having fun

Marketplace Dashboard

Route: /marketplace

Community dashboards, card presets, and themes:

Browse and install community-created dashboards (3+ available)
Card Presets for common use cases (7+ presets)
Theme marketplace with multiple visual styles (3+ themes)
CNCF project coverage tracker: 11 of 68 cards implemented (16%), with 35 Graduated, 33 Incubating, and 57 Help Wanted
Contributor Guide and Browse Issues links for community contribution
Rich tag-based filtering: alerts, argocd, certificates, clusters, compliance, cncf, cost, deployments, events, gitops, graduated, health, helm, incubating, monitoring, networking, observability, orchestration, pods, policies, production, provisioning, rbac, runtime, security, serverless, service-mesh, sre, storage, streaming, warm
Sort by Name, Type, or Author
Grid and list view toggle

Best for: Extending your console with community content

Utility Pages

These aren’t counted as dashboards but are useful:

Page	Route	What it does
Card History	`/history`	See cards you’ve removed
Settings	`/settings`	Configure your preferences
User Management	`/users`	Manage users (admin only)
Namespaces	`/namespaces`	Manage namespace access

Tips

Customizing Dashboards

Every dashboard can be customized:

Click “Add Card” to add new cards
Drag cards to rearrange them
Click the menu on any card to configure or remove it
Use the reset button to go back to defaults

Stats Blocks

The stats at the top of each dashboard show the most important numbers. You can configure which stats appear by clicking “Configure stats”.

Auto-Refresh

All dashboards auto-refresh by default. You can:

Toggle auto-refresh on/off
Manually refresh with the refresh button
See when data was last updated