Skip to content
Get packs

GKE Golden Base Image Finder

Finds the right GKE golden base image for custom node images from your cluster specs

Source Download skill
Repository
google/skills
Stars
21,782
Licence
Apache-2.0
Files
2 files

What it does

Google Kubernetes Engine (GKE) lets you build custom node images, which need a pre-approved 'golden' base image to start from. This skill teaches the agent to turn a plain-language description of the cluster you want (GKE version, OS, CPU architecture, GPUs or TPUs, gVisor sandboxing, cgroup mode) into the technical settings, then look up the matching base image name in Google's published image map. It also explains which settings it assumed when you did not specify them.

Great for

  • Platform or DevOps engineers preparing custom GKE node images
  • Engineers who know their workload needs (e.g. GPUs, ARM, gVisor) but not the exact image identifier
  • Teams checking whether a specific GKE version has a base image for their configuration

Skip it if

  • General GKE cluster creation or standard node pool management, which the skill itself says to hand off to other skills (gke-cluster-creation, gke-basics)
  • Anyone wanting a guaranteed-correct image without verifying it, since the skill infers defaults (e.g. COS, x86_64, cgroup v2) that may not match the real need and the image map is fetched live from the network

Try asking

  • “Find the golden base image for GKE 1.34.1-gke.2909000 on Ubuntu with ARM64 architecture”
  • “Which base image should I use for a custom GKE node with an H100 GPU?”
  • “I need a gVisor sandboxed GKE node image on COS for version 1.34, what base image do I use?”
Step-by-step guide: how to use GKE Golden Base Image FinderInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 2 original files, the Apache-2.0 licence and an attribution note.

    gke-custom-golden-image-discovery.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip gke-custom-golden-image-discovery.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip gke-custom-golden-image-discovery.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • State your GKE version explicitly; the skill requires it and rejects unsupported versions.
  • Mention hardware and security needs directly (GPU model, ARM, gVisor, hardened/signed modules) so the right flags are set rather than defaults.
  • Check the returned image name and the inferred settings before using them in Terraform or scripts, since the skill assumes defaults when you omit details.

What's inside

  • references/attribution.md
  • SKILL.md

Redistributed unmodified from google/skills under the Apache-2.0 licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.