Nuke

A Nushell-native Kubernetes toolkit

---

Overview

Have you ever done something like

kubectl get po | detect columns --guess
kubectl get po -o yaml | from yaml

hoping you could simply query and inspect data from the kube API-server?

Maybe you have something like this in your config

def kk [resource resourcename --namespace: string] {
    kubectl get $resource $resourcename --namespace $namespace | detect columns --guess 
    | update AGE {|p| $p.AGE | str replace s sec | str replace m min | str replace h hr | str replace d day | into datetime }
    | sort-by AGE
}

And you wish you could just kubectl get po | sort-by age? Me too.

That's why i created Nuke (that and to learn about kubernetes).
Nuke talks directly to the Kubernetes API server and returns structured, typed, queryable objects, so you can just

nuke get po | where restarts != 0 | sort-by restarts
nuke get po --all -l 'run in (production, quality)' -o wide | group-by node

Nuke tries to adhere the semantics of kubectl, integrating it with richer data. At the same time it allows you to create custom formatters for your favourite resources and commands.

Nuke does not aim to reimplement all of kubectl.
Instead, it focuses on those commands where Nushell’s structured data provides the most value.

All flags, resources and resource names support autocompletion.

---

Implemented Commands

Nuke Command	kubectl Equivalent
nuke get	kubectl get
nuke api-resources	kubectl api-resources
nuke api-versions	kubectl api-versions
nuke rollout status	kubectl rollout status
nuke rollout history	kubectl rollout history
nuke top	kubectl top
nuke config	kubectl config

---

Installation

Clone this repository into one of your Nushell library directories ($env.NU_LIB_DIRS):

let random_lib_dir = ([($env.NU_LIB_DIRS | first) nuke] | path join)
git clone git@github.com:lassoColombo/nuke.git $random_lib_dir

Verify installation:

use nuke
nuke api-resources

---

Output Formats

Commands that retrieve objects support three formats:

Format	Description
compact	Similar to kubectl get
wide	Similar to kubectl describe
full	Complete object returned by the Kubernetes API server

The compact format is the default when retrieving a list of objects, while wide is the default for single objects.

---

How Nuke Works

1. Reads your kubeconfig
2. Authenticates against the API server
3. Performs HTTP requests directly
4. Applies resource-specific formatter
5. Returns structured Nushell data

If no formatter exists, a default formatter is used.

---

Custom Formatters

Formatters control how Kubernetes objects are displayed.

You can override or define custom formatters using environment variables:

• NUKE_RESOURCE_FORMATTERS
• NUKE_ROLLOUT_FORMATTERS
• NUKE_METRIC_FORMATTERS

Example:

$env.NUKE_RESOURCE_FORMATTERS = {
  apps: {
    v1: {
      deployments: {|output?: string = compact|
        let obj = $in
        let res = {
          name: $obj.metadata.name
          namespace: $obj.metadata.namespace
          containers: ($obj.spec.template.spec.containers | length)
        }
        if $output == compact {
          return ($res
            | insert containers ($obj.spec.template.spec.containers | length)
          )
        }
        $res 
        | insert containers $obj.spec.template.spec.containers
      }
    }
  }
}

---

Config Module

Nuke provides utilities to manage kubeconfig contexts and namespaces.

Switch namespace

nuke config switch-namespace my-namespace
# If no argument is provided, you will be prompted to choose one in the builtin fuzzyfinder

Switch context

nuke config switch-context my-context
# If no argument is provided, you will be prompted to choose one in the builtin fuzzyfinder

Edit kubeconfig

nuke config edit

Additional helpers:

nuke config get-contexts
nuke config get-clusters
nuke config get-users
nuke config

---

Configuration

Nuke uses your existing Kubernetes configuration:

$env.KUBECONFIG  (defaults to ~/.kube/config)

No additional setup required.

Cache Location

Nuke follows the XDG Base Directory Specification:

$XDG_CACHE_HOME or ~/.cache

---

Authentication

Supported authentication methods:

• Token-based credentials
• Client certificate authentication

Planned support:

• OIDC
• Exec plugins

---

Dependencies

External tools required:

• curl — used for HTTP communication with the API server

Nuke otherwise aims to remain fully Nushell-native.

---

Updating

Pull the latest changes:

cd ([($env.NU_LIB_DIRS | first) nuke] | path join)
git pull

---

Contributing

Contributions, bug reports, and feature requests are welcome.

Before opening an issue or pull request, please read: CONTRIBUTING.md

The contributing guide includes:

• Development setup
• How to reproduce bugs
• KIND cluster configuration
• Metrics server setup
• Formatter development guidelines

---

Roadmap

• Improve coverage of built-in resource formatters
• Additional authentication methods
  • OIDC
  • Exec plugins
• Watch functionality