Skip to content
Get packs

GKE Workload Scaling

Guides an agent through configuring HPA and VPA autoscaling for GKE workloads, with best practices.

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

What it does

When installed, the agent follows built-in guidance for scaling applications on Google Kubernetes Engine: adding more or fewer pod replicas automatically (Horizontal Pod Autoscaler) or adjusting each pod's CPU and memory reservations (Vertical Pod Autoscaler). It can suggest kubectl and gcloud commands, provide YAML templates, and walk through rightsizing resource requests from observed usage. It also flags common pitfalls, such as using HPA and VPA on the same metric or auto-restarting pods without graceful shutdown handling.

Great for

  • Engineers running workloads on GKE who want to set up autoscaling correctly
  • Platform teams reviewing resource requests and limits across deployments
  • Teams trying to reduce overprovisioned CPU and memory spend on Kubernetes

Skip it if

  • Cluster-level node autoscaling (the skill explicitly excludes Cluster Autoscaler), since it does not cover that topic
  • Non-GKE Kubernetes or other clouds, where the gcloud steps and GKE-specific defaults would not apply
  • Running VPA in Auto mode on a single-replica production service without care, because the skill notes VPA restarts pods to change resources and may cause downtime

Try asking

  • “Set up a horizontal autoscaler for my checkout deployment that scales between 2 and 20 replicas on CPU”
  • “Enable VPA on my existing GKE Standard cluster and run it in Off mode to get recommendations”
  • “Help me rightsize the resource requests for my api deployment using VPA recommendations”
  • “Review my GKE deployment manifests for HPA and VPA conflicts and missing resource requests”
Step-by-step guide: how to use GKE Workload ScalingInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 4 original files, the Apache-2.0 licence and an attribution note.

    gke-workload-scaling.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip gke-workload-scaling.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip gke-workload-scaling.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

  • Start VPA in Off mode for at least 24 hours and review recommendations before applying any changes to requests.
  • Confirm the agent is targeting the correct cluster, namespace and deployment before running kubectl apply or gcloud update commands, since the skill's commands act on live infrastructure.
  • Make sure your containers handle SIGTERM and have Pod Disruption Budgets defined before enabling VPA Auto mode.

What's inside

  • assets/hpa-example.yaml
  • assets/vpa-example.yaml
  • 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.