GKE golden path: production-ready cluster defaults and readiness checklists
@google's opinionated GKE Autopilot configuration — production defaults, Day-0 vs Day-1 decisions, guardrails and readiness checklists
- What
- @google's opinionated GKE Autopilot configuration — production defaults, Day-0 vs Day-1 decisions, guardrails and readiness checklists
- Cost
- Free
- Needs
- a billed Google Cloud project with GKE access (clusters incur management fees and workload costs — no usable free tier); MCP tools or gcloud/kubectl configured; the skill itself is guidance only, it does not provision resources
- Install
- Copy the installer prompt below into your Muse — your agent does the rest.
Curated by Skill Harbor — 💳 Paid API required / API payante requise — GKE clusters run on a billed Google Cloud project (cluster management fees and workload resources are billed); the skill itself is guidance. @google's official golden-path configuration for production GKE clusters: an always-applied defaults table (Autopilot enabled, private nodes, private-endpoint enforcement, Secret Manager rotation, insecure RBAC bindings off, workload identity, ADVANCED_DATAPATH, Cloud DNS, Pod Security Standards `restricted` on production namespaces, shielded instances, stateful HA), a customer-configurable-settings table with deviation guidance (max pods per node, boot disk type, machine type, maintenance exclusions), production-readiness checklists, a Day-0 vs Day-1 framing that flags hard-to-change decisions (networking, private nodes, IP allocation), and guardrails — never request or output secrets, tool preference MCP > gcloud > kubectl, document every Day-0 choice and golden-path deviation. By @google, listed here with credit to its creator. Honest caveats: guidance and defaults only — it does not provision or change cloud resources itself; product defaults may drift from Google Cloud docs; requires GCP access for anything real. Apache-2.0 licensed. Updated October 20, 2026 (version 1.1.1): before any direct terminal command, the skill now points to references/attribution.md — every direct terminal `gcloud` invocation carries an inline usage-attribution tag (`agent-skills/1.0 (skill:gke-golden-path)`), and direct REST/MCP calls add it to the User-Agent — keep that prefix out of generated scripts, Terraform and user-facing code as the skill instructs; the tag identifies only the skill and its version, carries no user data, and a user opt-out is documented. Skill Harbor never reviews the code, review it yourself before use. Discovered via skills.sh.
Version: 1.1.1
Quick answer
GKE golden path is a free Muse skill, Google's official golden-path configuration for production GKE clusters, but the clusters run on a billed Google Cloud project, with cluster management fees and workload resources billed. It provides an always-applied defaults table for Autopilot, Day-0 versus Day-1 decisions, guardrails and readiness checklists for production clusters.
Install
Copy the install package below, then paste it into Muse🧪 Install ran in our test sandbox · Nov 18, 2026📌 Pinned version: commit 22bf5d1Step 0 — before installing anything: read this install prompt, summarize in a few lines what this build does and what installing it involves, then ask me to confirm before you proceed. Prerequisites: a billed Google Cloud project with GKE access (clusters incur management fees and workload costs — no usable free tier); MCP tools or gcloud/kubectl configured; the skill itself is guidance only, it does not provision resources Install "GKE golden path: production-ready cluster defaults and readiness checklists" for me. It gives the agent @google's official GKE golden-path configuration (version 1.1.1): the always-applied production defaults table (Autopilot enabled, private nodes, private-endpoint enforcement, Secret Manager rotation, insecure RBAC bindings off, workload identity, ADVANCED_DATAPATH, Cloud DNS, Pod Security Standards `restricted` on production namespaces, shielded instances, stateful HA), the customer-configurable-settings table with deviation guidance (max pods per node, boot disk type, machine type, maintenance exclusions), production-readiness checklists, the Day-0 vs Day-1 framing that flags hard-to-change decisions (networking, private nodes, IP allocation), the full cluster-level policy settings in assets/golden-path-autopilot.yaml, and the guardrails — never request or output secrets, tool preference MCP > gcloud > kubectl, document every Day-0 choice and golden-path deviation, and never promise zero downtime for node upgrades. Apache-2.0 licensed. Repository: https://raw.githubusercontent.com/google/skills/22bf5d15d546318171a1a27dda7fcdff9cb3029b/skills/cloud/gke-golden-path/SKILL.md (the skill is the whole skills/cloud/gke-golden-path/ folder at that commit: SKILL.md, assets/golden-path-autopilot.yaml, and references/attribution.md) Pinned version: install the code at commit 22bf5d15d546318171a1a27dda7fcdff9cb3029b exactly (Skill Harbor tested this version). Do not substitute a newer version. If this commit is unreachable, stop and tell me. [FINGERPRINTS] Files at the pinned commit, from Skill Harbor's review of 2026-10-20: - SKILL.md: SHA-256 7d2b3465899def03d1223395aca5a1b5c3695b1d38ea86540e95edcf7cddb2cb, 9,001 bytes, 8,989 characters - assets/golden-path-autopilot.yaml: SHA-256 7750ebbf07328f98674e4005ea9d7e81b4ba89d6b50cb6da18203cfbd9c18352, 4,776 bytes, 4,770 characters - references/attribution.md: SHA-256 0f985d6ec2cb3da41ee77174c9b5b17b6314435969bb80aebd02e555b0959e58, 1,885 bytes, 1,883 characters If any of these does not match, it changed since the review: do not install it yet, tell me what you found and let me decide. Updated October 20, 2026: Skill Harbor regenerated this install prompt from the current upstream file, at version 1.1.1 (the previous prompt described version 1.1.0). In SKILL.md, the only change besides the version is a new "Attribution" section: before executing any direct gcloud, REST API, or MCP command, the agent must read references/attribution.md; for commands executed directly in the terminal, prefix gcloud with CLOUDSDK_METRICS_REQUEST_ATTRIBUTION="agent-skills/1.0 (skill:gke-golden-path)" and append "agent-skills/1.0 (skill:gke-golden-path)" to the User-Agent of direct calls. That tagging applies only to direct terminal commands — never in generated scripts, Terraform, or the user's code — and it must not be persisted with `gcloud config set` (the prefix is inline, per command). As references/attribution.md states, the tag identifies only the skill and its version, it carries no user data, and a user opt-out is documented there: if the user asks to disable or remove tracking or attribution tags, omit the CLOUDSDK_METRICS_REQUEST_ATTRIBUTION variable and the custom User-Agent suffix from all commands. This prompt also now installs the complete folder explicitly — SKILL.md, assets/, and references/ — where the previous prompt mentioned SKILL.md and any helper files only vaguely. What is sent, and when: nothing is sent anywhere and nothing is executed during installation itself — installing only downloads the pinned folder from GitHub and copies it. The attribution tagging above applies later, when the skill is used and the agent runs direct terminal commands, as described in the safety check below. 1. Fetch the whole skills/cloud/gke-golden-path/ folder at the pinned commit into a temporary folder, then compare SKILL.md, assets/golden-path-autopilot.yaml, and references/attribution.md with the fingerprints from Skill Harbor's review of 2026-10-20 (see [FINGERPRINTS] above). If any does not match, do not install it yet, tell me what you found and let me decide. 2. Safety check: read SKILL.md, assets/golden-path-autopilot.yaml, and references/attribution.md and review them for anything suspicious (unexpected network calls, shell commands, credential harvesting); STOP on any red flag and tell me. Tell me in plain words what the skill does when used, which services it connects to, whether it reads or sends any personal information or keys, and anything that seems unusual. Confirm these points from the files: - The skill is guidance and defaults only — instructions, one YAML file of cluster-level policy settings, and the attribution reference. It ships no bundled scripts and no background service, and it does not provision or change cloud resources itself. Nothing is sent anywhere and nothing is executed during installation itself. Everything below happens in use, when I ask for GKE work, on my own billed Google Cloud project. - Guardrails, from SKILL.md: never request or output secrets (tokens, keys, service account JSON); this repo should contain zero secrets in code, credentials only via my own authenticated tools or the secure vault. Tool preference is MCP > gcloud > kubectl, respecting my choice if I say "use gcloud" or "use kubectl". Resolve project and cluster context from the conversation, MCP tools, or `gcloud config get-value project`, and ask me only if it cannot be resolved. Document decisions and rationale, especially Day-0 choices and golden-path deviations, and never promise zero downtime for node upgrades. - Day-0 vs Day-1: flag Day-0 decisions (networking, private nodes, subnets, IP allocation) prominently — they are hard or impossible to change after creation — and always ask clarifying questions before proceeding on them. For Day-1 features, propose the golden-path defaults with their trade-offs and let me confirm. The required inputs are project ID, region, and cluster name, plus environment type (dev/test or production, defaulting to production), networking (bring-your-own VPC/subnet or auto-create), scale expectations, and cost constraints. - Attribution, the change in this version: before executing any gcloud, REST API, or MCP command, read references/attribution.md. For commands executed directly in the terminal only — never in generated scripts, Terraform, or user code — prefix gcloud with CLOUDSDK_METRICS_REQUEST_ATTRIBUTION="agent-skills/1.0 (skill:gke-golden-path)", set inline on each command, and append "agent-skills/1.0 (skill:gke-golden-path)" to the User-Agent header of direct HTTP, REST, or MCP requests (appending to an existing User-Agent rather than overwriting it). Do not persist the tag with `gcloud config set` — it would outlive the task and mislabel unrelated usage. The tag identifies only the skill and its version and carries no user data. If I ask to disable or remove tracking or attribution tags, omit the variable and the User-Agent suffix from all commands, as the documented opt-out in references/attribution.md provides. 3. When I say "yes, install", install it as a skill: copy the whole fetched skills/cloud/gke-golden-path/ folder (SKILL.md, assets/, and references/, keeping the subfolders so the relative links in SKILL.md keep working) into the agent's skills directory, in a folder named "google-skills-gke-golden-path". 4. Verify with no network calls: frontmatter valid, files in place, and the copied SKILL.md still matches its fingerprint above. 5. Report what was installed, where, and what I still need to do myself (e.g. gather the required inputs — project ID, region, cluster name, environment type — and confirm the Day-0 decisions before any real provisioning; no GKE resource is created and no billed cluster change happens until I ask for GKE work). GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder and the install target. If anything looks off, stop and ask me.
Saved to your recent installs. Find it anytime on /connect.
🔒 Security, provenance and what this touches
The install prompt below already includes the vetting steps: your agent follows the community checklist before installing anything with executable code. Want more?
Verification certificate
Dated facts about what was checked for this listing, and how. This is not a safety certification.
- Pinned version
- 22bf5d1 · 2026-10-20
- Sandbox install test
- Install ran in our test sandbox · Nov 18, 2026
- Creator identity
- Not verified
- Fingerprint watch
- Last checked · Nov 19, 2026 : no change detected in the files linked by this install prompt
- Install prompt analysis
- 2026-10-13 : No commands, network calls, packages, file writes or API keys detected.
Skill Harbor never reviews the code of a listed skill. These are dated facts about the install prompt and our install test, not a safety certification. Review a skill yourself before you install it. How we verify
Questions
How do I install a skill?
Every product page includes a copy-paste install prompt. Paste it into your Muse and it sets the skill up for you — no manual configuration.
Where does my money go?
Straight to the seller. Skill Harbor never processes payments: checkout happens on the seller’s own page, usually Stripe.
What does the ✓ next to a creator’s name mean?
It means we confirmed the identity of the person behind the listing. It says nothing about the code itself — always check a skill before installing it.