What does this skill do, and when should you use it?
This skill is an entry point for Google Cloud’s managed Kubernetes platform, GKE, with Autopilot as its default mode. It provides quick-start commands for enabling the Kubernetes Engine API, creating an Autopilot cluster, and retrieving credentials. It routes concrete workloads to one most-specific GKE sub-skill rather than handling specialized areas itself. For purely conceptual questions, it directs the agent to answer immediately without tool calls.
It can guide an agent through enabling the Kubernetes Engine API, creating an Autopilot cluster with gcloud, and retrieving cluster credentials. It also maps request keywords such as cluster creation, networking, Workload Identity, autoscaling, storage, upgrades, and observability to named GKE specialist skills. For conceptual questions such as what GKE is or generic Standard-versus-Autopilot comparisons, it directs the agent to answer from existing knowledge.
- A developer or platform engineer who needs a quick Autopilot GKE cluster on Google Cloud.
- A user whose GKE request concerns a specialist area such as Ingress, Workload Identity, scaling, storage, or upgrades.
- A platform team that wants an agent to classify GKE requests before handing them to one focused workflow.
- A learner asking conceptual GKE questions without wanting code searches or other tool calls.
How do you install this skill?
- Quick Start creates real cloud resources and may incur charges; confirm project, region, permissions, and budget before execution, and add dry-run or rollback guidance instead of relying on --quiet.
- Do not default to roles/container.admin; check authorization and use the minimum IAM roles required for each operation.
- MCP tools, gcloud versions, GKE API fields, and example commands are not supported by tests or CI in the supplied material; version pinning and regression testing are needed before production use.
- Chinese interaction and mainland-China network reachability are not documented; dependent Google Cloud/MCP services may limit practical use for FollowSkills users in China.
- Shell / CLI
- Network access
Google Cloud gcloud CLI
In an Agent Skills-compatible environment, run: npx skills add google/skills. Select the specific skill during installation; this skill is located at skills/cloud/gke-basics. The source does not document other installation paths or client-specific configuration.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/google/skills.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/skills/cloud/gke-basics" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- Create an Autopilot GKE cluster in us-central1
- I need to configure GKE Ingress.
After installation, try a prompt such as “Create an Autopilot GKE cluster in us-central1” or “I need to configure GKE Ingress.” The documented starter commands are: gcloud services enable container.googleapis.com --quiet; gcloud container clusters create-auto my-cluster --region=us-central1 --quiet; gcloud container clusters get-credentials my-cluster --region=us-central1 --quiet. For specialist work, the agent should load the single most-specific sub-skill from the routing table.
What are this skill's strengths and limitations?
- Clear routing table covering GKE creation, networking, security, scaling, storage, upgrades, observability, and other scenarios.
- Includes copyable commands for a basic Autopilot cluster workflow.
- Defines a clear direct-answer path for conceptual questions.
- Distributed under the Apache-2.0 license in an actively developed repository.
- It does not itself provide detailed implementation guidance for specialist areas such as networking or security.
- No test suite, compatibility matrix, or failure-recovery documentation is provided in the source.
- The cluster workflow requires the gcloud CLI and appropriate Google Cloud permissions; exact permissions are not documented.
- The single-skill routing rule may require additional coordination for genuinely cross-domain tasks.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| GKE Basics Navigator this page ✓ Google · Official | 45 · Not recommended | ★ 21k | 12d ago | Apache-2.0 |
| GKE Production Golden Path ✓ Google · Official | 54 · Use with care | ★ 21k | 12d ago | Apache-2.0 |
| GKE Cluster Creation Advisor ✓ Google · Official | 48 · Use with care | ★ 21k | 12d ago | Apache-2.0 |
| AI Workload Migration to GKE Inference ✓ Google · Official | 50 · Use with care | ★ 21k | 12d ago | Apache-2.0 |
| GKE AI Inference Deployment Assistant ✓ Google · Official | 46 · Use with care | ★ 21k | 12d ago | Apache-2.0 |
It defaults to GKE Autopilot. For generic Standard-versus-Autopilot questions, it directs the agent to answer conceptually, but it does not provide a full comparison table or migration guide.
How did FollowSkills review this skill?
The material shows MCP-first guidance, dry-run support, explicit READ/MUTATE/DESTRUCTIVE labels, and security concepts such as Workload Identity and Secret Manager. However, Quick Start directly enables a service and creates a cluster with --quiet, without confirmation; credential, log, and cluster-data flows, rollback, and least-privilege guidance are incomplete, and the troubleshooting text suggests the broad container.admin role. Points are deducted accordingly.
The documentation covers the main tool paths, fallback order, and some error remediation, making the happy path plausible. Reliability is reduced by malformed tables, placeholder parameters, unverified MCP availability, client examples that ignore errors, and thin handling of abnormal inputs and failure feedback. Static calibration caps this dimension at10, so6 is awarded.
The skill clearly acts as a GKE routing hub, with scenarios, trigger keywords, specialized-skill boundaries, and a rule for conceptual questions. It does not sufficiently state non-fit ranges, environment prerequisites, false-trigger boundaries, or Chinese-language support. Core operation depends on Google Cloud, GKE MCP, gcloud, or kubectl, with no mainland-China reachability discussion, so points are deducted.
The material has useful layering through quick start, routing, reference files, tool preferences, error tables, installation guidance, repository provenance, and an Apache-2.0 license. It lacks skill-level versioning, a changelog, a named maintenance owner, and a clear update path; permission and credential assumptions remain implicit, and malformed tables hurt maintainability. Points are deducted for these gaps.
The skill provides clear routing for GKE work and useful references for MCP, gcloud, kubectl, client libraries, and Terraform, so its navigation value is real. It does not itself complete most specialized tasks, the quick start is not directly production-ready, and examples and current API/tool compatibility are not verified. Static calibration caps this dimension at7, so5 is awarded.
Commands, code samples, tool mappings, symptoms, remediations, and official-document references provide limited auditability. There is no committed test suite, CI coverage, or third-party execution evidence in the supplied material; MCP tool count and behavior are primarily author assertions and cannot be independently reproduced from a static read. Static calibration caps this dimension at5, so3 is awarded.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →