k13d

module
v0.9.2 Latest Latest
Warning

This package is not in the latest version of its module.

Go to latest
Published: Feb 19, 2026 License: MIT

README

k13d

The all-in-one Kubernetes dashboard — Terminal & Web UI with AI built in.

kubeaidashboard = k+ 13 letters + d = k13d

Download a single binary, run one command, and get a full-featured Kubernetes dashboard
with an AI assistant that actually executes commands for you.

Release Go K8s License AI

Documentation · Download · Web UI Guide · TUI Guide · 한국어


Web UI

Web UI Dashboard

TUI

TUI Dashboard


Get Started in 30 Seconds

1. Download from Releases — single binary, no dependencies.

tar xzf k13d_*.tar.gz && chmod +x k13d

macOS: if blocked, run xattr -d com.apple.quarantine ./k13d

2. Run.

TUI mode — terminal dashboard with Vim navigation + AI panel
./k13d

Opens a full-featured terminal dashboard. Use j/k to navigate, Tab to open the AI panel, : for commands.

Web UI mode — browser dashboard
./k13d -web -auth-mode local
# Open http://localhost:8080 — Default login: admin / admin

Starts an HTTP server with the web dashboard. Add -port 3000 to change the port.

Both modes — Web UI + TUI simultaneously
./k13d -web -auth-mode local &   # Web UI in background
./k13d                            # TUI in foreground

Run the Web UI as a background process, then launch the TUI in the foreground. Both share the same kubeconfig.

Flags
Flag Description
-web Start Web UI server (default port 8080)
-port <N> Custom port for Web UI
-auth-mode local Enable local username/password auth
--no-auth Disable auth (dev only)
--embedded-llm Use built-in LLM (no API key needed)
--kubeconfig <path> Custom kubeconfig path
--context <name> Use specific cluster context
--debug Enable debug logging

That's it. Your kubeconfig is auto-detected.


Why k13d?

k13d k9s Lens kubectl
Terminal UI Yes Yes - -
Web UI Yes - Yes -
AI Assistant Yes - - -
Single binary, zero deps Yes Yes - Yes
Free & open source Yes Yes Paid Yes

Web UI — Everything in the browser

  • Dashboard — Pods, Deployments, Services, all resources with real-time status
  • AI Assistant — Ask questions, AI executes kubectl with your approval
  • Topology — Graph & tree visualization of resource relationships
  • Reports — Cluster health, security audit, FinOps cost analysis
  • Metrics — Historical CPU/Memory/Pods/Nodes charts (SQLite-backed)
  • Helm — Release management, history, rollback
  • Terminal — Full xterm.js shell into any pod
  • Logs — Real-time streaming with ANSI colors, search, download
  • RBAC Viewer — Subject-to-role relationship map
  • Network Policy Map — Ingress/egress rule visualization
  • Event Timeline — Cluster events grouped by time windows
  • Resource Templates — One-click deploy (Nginx, Redis, PostgreSQL, etc.)
  • Notifications — Slack, Discord, Teams, Email alerts
  • 5 Themes — Light (default), Tokyo Night, Production, Staging, Development

TUI — k9s on steroids

  • Vim navigationj/k, g/G, / filter, : commands
  • AI panelTab to chat, AI executes commands for you
  • SortShift+N name, Shift+A age, Shift+T status, :sort picker
  • Autocomplete — Dropdown suggestions as you type
  • Aliases — Custom shortcuts (pp -> pods)
  • Plugins — External tool integration via plugins.yaml
  • i18n — English, Korean, Chinese, Japanese

AI Setup (Optional)

Configure in Settings > AI in the Web UI, or via environment:

# OpenAI
export OPENAI_API_KEY=sk-...
./k13d -web -auth-mode local

# Ollama (local, free, no API key)
ollama pull qwen2.5:3b && ollama serve
./k13d -web -auth-mode local
# Set Provider: "ollama" in Settings > AI

The AI assistant can:

  • Diagnose pod crashes and suggest fixes
  • Execute kubectl commands with your approval
  • Scale deployments, restart rollouts
  • Analyze YAML, events, and logs in context

CLI Reference

./k13d                              # TUI mode
./k13d -web                         # Web UI (port 8080)
./k13d -web -port 3000              # Custom port
./k13d -web -auth-mode local        # With authentication
./k13d -web --no-auth               # No auth (dev only)
./k13d --kubeconfig ~/.kube/prod    # Custom kubeconfig
./k13d --context prod-cluster       # Specific context
./k13d --debug                      # Debug logging

Docker

docker run -d -p 8080:8080 \
  -v ~/.kube/config:/home/k13d/.kube/config:ro \
  cloudbro/k13d:latest \
  -web -auth-mode local

Build from Source

git clone https://github.com/cloudbro-kube-ai/k13d.git && cd k13d
make build

Documentation

Full documentation: https://cloudbro-kube-ai.github.io/k13d


License

MIT License - see LICENSE.

GitHub Stars

Directories

Path Synopsis
cmd
bench command
Package main provides the CLI for running AI benchmarks
Package main provides the CLI for running AI benchmarks
eval command
Package main provides the CLI for running LLM evaluation benchmarks.
Package main provides the CLI for running LLM evaluation benchmarks.
kube-ai-dashboard-cli command
TODO: This file duplicates significant logic from cmd/kubectl-k13d/main.go.
TODO: This file duplicates significant logic from cmd/kubectl-k13d/main.go.
kubectl-k13d command
kubectl-k13d is a kubectl plugin wrapper for k13d.
kubectl-k13d is a kubectl plugin wrapper for k13d.
pkg
ai
ai/agent
Package agent provides a state-machine based AI agent following kubectl-ai patterns.
Package agent provides a state-machine based AI agent following kubectl-ai patterns.
ai/safety
Package safety provides command safety analysis using shell AST parsing.
Package safety provides command safety analysis using shell AST parsing.
ai/session
Package session provides AI conversation session persistence
Package session provides AI conversation session persistence
ai/sessions
Package sessions provides the original session management system.
Package sessions provides the original session management system.
bench
Package bench provides AI benchmarking capabilities for Kubernetes tasks.
Package bench provides AI benchmarking capabilities for Kubernetes tasks.
bench/cluster
Package cluster provides cluster provisioning for benchmarks
Package cluster provides cluster provisioning for benchmarks
db
eval
Package eval provides AI agent benchmark evaluation framework dryrun.go implements cluster-free benchmark evaluation using tool call analysis
Package eval provides AI agent benchmark evaluation framework dryrun.go implements cluster-free benchmark evaluation using tool call analysis
k8s
llm/embedded
Package embedded provides an embedded LLM server using llama.cpp This is an optional component that can run a local SLLM for environments without external LLM access.
Package embedded provides an embedded LLM server using llama.cpp This is an optional component that can run a local SLLM for environments without external LLM access.
log
mcp
metrics
Package metrics provides periodic metrics collection for time-series data
Package metrics provides periodic metrics collection for time-series data
security
Package security provides security scanning and compliance checking for Kubernetes clusters
Package security provides security scanning and compliance checking for Kubernetes clusters
testutil
Package testutil provides shared testing utilities and interface contracts.
Package testutil provides shared testing utilities and interface contracts.
ui
Package ui provides the terminal user interface components.
Package ui provides the terminal user interface components.
ui/actions
Package actions provides a centralized key action system following k9s patterns.
Package actions provides a centralized key action system following k9s patterns.
ui/models
Package models provides data models for the TUI following k9s patterns.
Package models provides data models for the TUI following k9s patterns.
ui/render
Package render provides resource rendering following k9s patterns.
Package render provides resource rendering following k9s patterns.
ui/resources
Package resources provides resource view generators for the TUI dashboard.
Package resources provides resource view generators for the TUI dashboard.
ui/views
Package views provides high-level view components following k9s patterns.
Package views provides high-level view components following k9s patterns.
web
tests
mocks command
Mock OpenAI API Server for Integration Testing Simulates OpenAI API responses for testing k13d without real API calls
Mock OpenAI API Server for Integration Testing Simulates OpenAI API responses for testing k13d without real API calls

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL