p10node/k10s

The Kubernetes terminal UI you can click.

k9s taught us to live in the terminal. k10s makes that terminal point-and-click, instantly searchable, themeable - and gives it an AI that already knows your cluster, namespace and selected object.

Install · Try it with no cluster · Features · k9s → k10s · Docs · Build log

_{A real terminal capture of k10s demo running - just screenshot, on the offline demo backend. Every frame in these docs comes out of the running binary, never hand-drawn.}

A cluster dashboard is something you open twenty times a day, usually while something is on fire. The two things that matter are how fast it opens and how little you have to remember.

• Nothing is hidden behind memorised keys. The actions that apply to the thing you selected are listed, right there, in their own pane. Click one, or press the letter next to it.
• Your mouse works. Click a row, click a pane, click [ zoom ], click ns default ▾, scroll the table. Every border button is real.
• One search box for the whole cluster. ctrl+p searches resource kinds and objects together. No prefix language to learn.
• It opens instantly. Startup registers zero watches and waits for nothing - informers start lazily, per kind, the first time you look at one. Opening Pods watches pods, not every Secret and Event you own. (how, and the regression guards)
• AI that can see the screen. ctrl+a, ask in plain English. The current context, namespace, kind and selected object are injected into the prompt, so "why is this pod unhealthy?" means this pod.
• It updates itself. /update installs the newest release over the running binary - checksum-verified, atomic, offers to restart into it.
• Your k9s-style command plugins fit. Put scoped shortcuts in ~/.k10s/plugins.yaml; they appear beside built-in actions and receive the selected object, namespace, context and column values.

k10s ships an offline demo backend: a realistic cluster to click around in, including a CrashLoopBackOff to poke at. It is opt-in, because sample data should never be mistaken for your machine.

git clone https://github.com/p10node/k10s && cd k10s
go run . demo     # the sample cluster - fake data, clearly labelled
go run .          # your real cluster, or "No cluster" if there isn't one

The demo is a context, not a mode. k10s demo opens on it, /demo switches to it from anywhere, and :ctx always lists it (labelled k10s demo · sample data, with a legend under the list). To leave it, pick any other context - there is no separate exit. While it is up the header carries a DEMO marker, so no frame of it can be mistaken for a real cluster.

Plain k10s reads the same kubeconfig kubectl does ($KUBECONFIG, else ~/.kube/config) and shows what that context can reach - and nothing else. With no kubeconfig, or a context whose API server does not answer, the main panel says No cluster and points at the way in: r retries, :ctx picks another context, /setup has the kubectl and kubeconfig links. docs/cluster-setup.md is the same guide, longer.

curl -fsSL https://p10node.com/k10s/install.sh | sh

macOS and Linux, amd64 and arm64. It picks the right prebuilt binary, verifies its sha256 against the release manifest, and installs it into /usr/local/bin (or ~/.local/bin when that needs a password it cannot ask for). --dir, --version and --no-sudo are in docs/install.md, along with how to read it before you run it and the matching uninstall.sh.

With Go on the box:

go install github.com/p10node/k10s@latest

Or from a clone, which also stamps the version into the binary:

git clone https://github.com/p10node/k10s && cd k10s
just install                 # → $GOBIN/k10s, version-stamped

Prebuilt static binaries for darwin/amd64, darwin/arm64, linux/amd64, linux/arm64 and windows/amd64 are published on every tag — Windows is the one platform the installer script sends to the release page instead. Every later upgrade is just:

k10s update

Then run it:

k10s                  # your current kubeconfig context
k10s demo             # the built-in sample cluster, no cluster needed
k10s --readonly       # look, never touch: nothing that changes the cluster
k10s --version        # which build is this

30 kinds across Workloads · Network · Config · Storage · RBAC · Cluster · Custom Resources, each with a live row count for the current namespace. Your CRDs are discovered automatically - no configuration.

Groups fold: space, left or a click on the header, remembered across restarts, with Config/Storage/RBAC folded to begin with. A folded group also asks your cluster for nothing - the sidebar is what decides which kinds get counted, and counting is one limit=1 request per visible kind, six at a time, with refusals remembered.

Pods · Deployments · ReplicaSets · StatefulSets · DaemonSets · Jobs · CronJobs · HPAs · Services · Endpoints · Ingresses · NetworkPolicies · ConfigMaps · Secrets · ResourceQuotas · LimitRanges · PDBs · PVCs · PVs · StorageClasses · ServiceAccounts · Roles · RoleBindings · ClusterRoles · ClusterRoleBindings · Nodes · Namespaces · Events · CRDs · Custom Resources

Add your own scoped actions with the core k9s plugins.yaml format. They can run foreground or background commands, request confirmation, override a built-in shortcut deliberately, and are clickable in the same pane. See the plugin guide and installable example.

╭─ logs -f billing-worker-6f8d9c5b7-qq91x ────────────────────────────────── [ close ] [ restore ] ╮
│   8 2026-08-25T08:12:18.331Z INFO  http          GET  /v1/users/me         200 3.1ms             │
│     trace=44b1e2f9                                                                               │
│   7 2026-08-25T08:12:21.660Z INFO  worker        flushed batch size=250 dur=41ms                 │
│   6 2026-08-25T08:12:25.019Z INFO  http          GET  /healthz             200 0.3ms             │
│   5 2026-08-25T08:12:31.402Z INFO  http          DELETE /v1/sessions/9a1   204 5.7ms             │
│     trace=7c0d19ba                                                                               │
│   4 2026-08-25T08:12:33.881Z WARN  gc            pause=18ms heap=412Mi                           │
│   3 2026-08-25T08:12:40.117Z INFO  http          GET  /v1/orders/88213     200 7.4ms             │
│     trace=e21f8b05                                                                               │
│   2 2026-08-25T08:12:44.590Z INFO  metrics       scrape ok series=1842                           │
│   1 2026-08-25T08:12:51.008Z INFO  http          GET  /healthz             200 0.3ms             │
│ ● following  newest at bottom                                        500 loaded · ↑ for older    │
╰──────────────────────────────────────────────────────────────────────────────────────────────────╯

╭─ Prompt · plain text → AI · /commands still work · esc close ─ [ grow ] [ AI · claude-sonnet-5 ] ╮
│ ✦ ask about your cluster…   ·   /settings to change provider/model                               │
╰──────────────────────────────────────────────────────────────────────────────────────────────────╯
 ● AI mode — plain text goes to claude-sonnet-5    tab panes · enter open · ctrl+p search · f find…

ctrl+a toggles it. Bring your own key - OpenAI-compatible (so also Groq, Together, OpenRouter, vLLM, Ollama, LM Studio, any local gateway) or Anthropic. Answers open as a normal text view you can scroll, zoom and close. No cluster data leaves your machine unless you press enter in AI mode. The only other network call k10s makes on its own is the once-a-day update check, which asks GitHub for a version number and nothing else.

tokyo-night (default) · catppuccin-mocha · dracula · nord · gruvbox-dark · solarized-dark · solarized-light · matrix

T cycles, /theme opens a picker that applies each theme as you move through the list - you judge it on the real UI, not on a name - and esc puts back whatever you had. Drop a YAML palette into ~/.k10s/themes, restart, and it appears in the same picker — no rebuild needed. See the custom-theme guide and installable demo.

• Copy mode (ctrl+s) releases the mouse so your terminal can drag-select and copy - the one thing every mouse-capturing TUI breaks.
• An honest loading state instead of "no resources found" while a watch is still syncing.
• Namespace and context pickers that never ask you to type a name.
• No setup screen. First run opens the cluster; every setting has a working default and /settings is one keystroke away when you want it.
• A grow-able prompt (ctrl+z) - because a long kubectl line or an AI question does not fit in a one-row field that scrolls sideways.
• A terminal-too-small notice instead of a garbled layout.

k10s exists because of k9s. Credit where it is due: k9s is mature, enormous in scope, plugin-extensible, and it is the reason a whole generation of us stopped typing kubectl get pods all day.

k10s is younger and deliberately narrower. The difference is philosophy:

Use k9s if you want the biggest feature surface, its full plugin ecosystem, and years of production mileage. Use k10s if you want something you can hand to a teammate who has never opened a TUI, and have them find logs on their own in ten seconds.

The short version - the full reference is here.

Two prefixes, and each one only ever shows its own set. / is k10s itself (theme, settings, updates). : is the cluster, in the k9s vocabulary - a resource view, the namespace, the context, or something acting on what is on screen. enter runs the highlighted suggestion immediately - no second trip through the prompt.

/theme  /settings  /mouse  /update  /version  /demo  /setup  /help    k10s itself

:po  :deploy  :rs  :sts  :ds  :job  :cj  :hpa      workloads
:svc  :ep  :ing  :netpol                           network
:cm  :sec  :quota  :limits  :pdb                   config
:pvc  :pv  :sc                                     storage
:sa  :role  :rb  :crole  :crb                      rbac
:no  :ns  :ev  :crd  :cr                           cluster + custom

the short form is what the popup lists; every kind also answers to its
plural and singular - :pods, :pod, :deployments, :persistentvolumes ...
and `:aliases` prints the lot

:po kube-system                 that kind, in that namespace (`all` works)
:svc api                        an argument that is not a namespace filters
:ns   :ctx                      namespace / context picker
:ns <name>   :ctx <name>        switch either outright
:aliases  :search <term>  :filter <term>  :scale <n>  :q

/ is k10s, : is the cluster - namespace and context are :ns and :ctx, not /ns and /context. :aliases prints the whole : vocabulary.

Anything that is not a / or : command runs as a shell command and its output opens in the main panel - date, kubectl get pods -o wide, helm list. Run through your own $SHELL, capped at 30s, with no terminal attached (use s on a pod for an interactive shell). Full command reference.

k10s opens on kubeconfig's current-context — the cluster kubectl would talk to. :ctx <name> switches for the session without touching kubeconfig; kubectl config use-context is what changes where it opens next time.

Theme, namespace, folded sidebar groups, AI settings and the update check persist to ~/.k10s/config.yaml (override with K10S_CONFIG), saved on every change. See config.md.

Note: the AI API key is stored as plain text in that file (mode 0600) and masked in the UI. If that is not acceptable in your environment, leave it empty and AI mode simply stays off.

The barrier is low on purpose. just lists every recipe:

just dev            # run from source
just shot 140 44    # render one frame headlessly - no TTY, no cluster
just screenshot     # capture the README hero from a real terminal (vhs)
just check          # fmt + vet + test, in the order that fails fastest

cmd/shot is why UI work here is pleasant: you can iterate on layout, replay keystrokes (just shot 140 44 j,j,d) and diff frames without a terminal or a cluster. Nothing in these docs is drawn by hand - the frames come from that renderer, the hero image from just screenshot driving the real binary in a real terminal.

Good first issues live in roadmap.md → possible next steps: more kinds, click-to-sort columns, multi-select bulk delete, grep inside the log stream, saved views.

Honest status: every live path is implemented against the real client-go/kubectl APIs and covered by tests with fake clientsets, but this tree has not yet been smoke-tested against a production cluster by its author. Try it on kind first. The known limits are listed in the roadmap rather than buried.

To my colleagues - the people whose questions, over-the-shoulder debugging sessions and "can you just check if that pod is up?" pings are the entire reason this exists. k10s was not built to be a project; it was built so that looking at our clusters would stop being the annoying part of your day. Every pane in it is an answer to something one of you asked me. Thank you for the inspiration, the patience, and for being the first users.

Cảm ơn các đồng nghiệp của tôi - repo này được viết ra để phục vụ các bạn, và chính các bạn là nguồn cảm hứng cho k10s.

To @derailed and k9s, for proving a Kubernetes TUI could be something you actually want to open.

To Charm - Bubble Tea, Lip Gloss, Bubbles - and to BubbleZone, which is the reason any of this is clickable at all.

Apache-2.0 - the same license as Kubernetes, Helm and k9s. Use it, fork it, ship it inside your company.