MSKazemi/kubeintellect v2.3.0

v2.3.0 — distribution channels

on GitHub

Distribution channels

This release is mostly about how you get KubeIntellect. Three new channels, all generated from this repository on every tag.

Install

# Server, into a cluster
helm install kubeintellect oci://ghcr.io/mskazemi/charts/kubeintellect \
  --namespace kubeintellect --create-namespace

# Terminal client
brew tap MSKazemi/kube-q && brew install kube-q     # macOS / Linux
pipx install kube-q                                  # any OS with Python

Standalone kq binaries for linux and darwin on amd64 and arm64 are attached below with checksums — they need no Python on the target machine.

Added

• The Helm chart is published — oci://ghcr.io/mskazemi/charts/kubeintellect. It existed in-tree but had never been pushed anywhere, so there was no supported way to install the server. It is listed on Artifact Hub.
• Standalone kq binaries on every release, frozen with PyInstaller.
• kubectl kq submitted to krew, so the client is installable as a kubectl plugin.

Changed

• The Homebrew formula moved to its own tap, MSKazemi/homebrew-kube-q. The install docs had instructed brew tap MSKazemi/kube-q in four places for a tap that did not exist. Its CI now runs brew audit --strict --online and actually builds and runs the formula on every push.
• Three new playbooks and worked examples for eight kq subcommands landed since v2.2.0 — the library is now 23 playbooks, 20 of which compile to zero-token detectors.

Fixed

• The Helm chart could never pull its image. values.yaml defaulted image.tag to dev-latest, a tag that has never existed in the registry, so a default helm install went straight to ImagePullBackOff. It now defaults to the chart's appVersion, with a CI assertion so it cannot regress.
• The container image had never reached Docker Hub. docker.io/kazemi/kubeintellect now exists.
• Two brew audit --strict violations that the earlier formula fix missed — invisible because brew had never once been run against it.

Full detail in CHANGELOG.md.