# kubernetes/examples: what the official SIG Apps repo actually contains > The kubernetes/examples repository collects deployable Kubernetes sample manifests under one SIG Apps umbrella, with AI, web, database and platform-app directories. It is a reading and teaching resource, not a package you install, and the per-example READMEs carry the real instructions. **kubernetes/examples** — Kubernetes application example tutorials - Repository: https://github.com/kubernetes/examples - Stars: 6,771 · Forks: 4,747 - Language: Shell - License: Apache-2.0 - Published: 2026-09-22 · Updated: 2026-09-22 · Language: en - Canonical page: https://hysenlabs.com/projects/kubernetes-examples ## What kubernetes/examples is, and what it is not This is a curated collection of Kubernetes application example tutorials, stewarded by SIG Apps and published in the kubernetes organization. The README frames the goal in three verbs: discover practical implementations, learn common patterns, and explore configurations for general-purpose applications plus specialized ones such as AI/ML workloads and platforms. The repository is written mostly in Shell, licensed under Apache-2.0, and its default branch is master. The last push recorded for it was 2026-03-03. What it is not matters more for adoption. There is no homepage, no release list, and no published artifact. Nothing here installs as a package. You clone or browse the tree, pick a subdirectory, and follow that subdirectory's own README. The top-level README is an index and a contribution guide, not a manual. If you arrive expecting a Helm repository or a versioned bundle, you will not find one, and the README does not claim otherwise. ## How the repository is laid out, and why the layout is the product The top level is deliberately shallow. Alongside README.md, LICENSE, CONTRIBUTING.md, guidelines.md, OWNERS, SECURITY.md, SECURITY_CONTACTS, code-of-conduct.md and .gitignore, the working directories are AI/, databases/, nginx-platform-app/ and web/. There is also an _archived/ directory, which is where examples that no longer meet the bar appear to be moved rather than deleted. That structure is the mechanism. Each category is a folder of self-contained examples, and each example carries its own README.md covering purpose and learning objectives, prerequisites including Kubernetes version and hardware where applicable, step-by-step deployment instructions, verification and cleanup commands, and the key Kubernetes concepts used. So the data flow is: you enter at the top-level README, it points you at a category, the category points you at an example, and the example README is the actual tutorial. Nothing links these together at runtime. There is no controller, no operator, no shared library. The only shared thing is the guidelines.md standard that contributors must follow, which is what keeps the per-example READMEs consistent enough to be usable. ## Cloning the repository and running a first example There is no install step because there is no software to install. The README gives no installation instructions for the repository itself; it directs you to the subdirectories. The practical first move is to get a local copy, using the clone URL the repository is published at. ```bash git clone https://github.com/kubernetes/examples.git cd examples ls ``` After that you should see the top-level entries the README describes: AI/, databases/, nginx-platform-app/, web/ and the supporting documents. From there, pick a category and read its contents before applying anything. The README states that each example's README covers prerequisites including Kubernetes version and hardware if applicable, so that file is where you confirm your cluster is suitable. Only then do you apply the manifests, using the exact commands that example's README provides. The top-level README does not supply a kubectl apply line of its own, and writing one here would be guessing at a directory layout the README does not spell out. Treat the per-example README as the only authoritative source of commands, ports and configuration keys. ## The _archived directory is the honest signal about staleness Kubernetes moves faster than any example collection can. API versions graduate and get removed, admission behaviour changes, and a manifest that applied cleanly two years ago can fail validation today. The repository's answer to this is _archived/. Examples that no longer meet the guidelines appear to be moved there instead of being deleted, which preserves them for reading while separating them from the maintained set. This is a better arrangement than silent rot, but it is not a guarantee. The top-level README's philosophy section says examples should be current and maintained, yet nothing in the top-level README tells you how recently any individual example was touched. The last push to the repository as a whole was 2026-03-03, and that single date says nothing about whether the example you picked was edited in that push or three years earlier. If currency matters for your use, check the per-directory history rather than the repository-level date. The README does not document a deprecation policy beyond the existence of _archived/, and it does not document rollback. ## Where this is the wrong tool kubernetes/examples is the wrong tool when you need something that runs in production under a support commitment. Nothing here is versioned, nothing is released, and there is no compatibility matrix. A sample manifest that demonstrates a pattern is not a hardened deployment: the README itself describes the examples as real-world but possibly simplified. It is also the wrong tool for automated consumption. If your pipeline needs to pull a chart with a pinned version and a values schema, this repository offers neither. And it is the wrong first stop for someone who does not yet have a working cluster. The examples assume you can apply manifests and inspect the result; the per-example READMEs state prerequisites rather than teaching you to install Kubernetes. Finally, if you need a single canonical answer for how to run a specific piece of software, a project's own documentation will beat a community example every time, because the upstream project owns the version compatibility story and this repository does not. ## How it compares with the alternatives engineers actually reach for The closest alternative in spirit is the tutorials section of the Kubernetes documentation itself. That material is prose-first: it walks you through a concept and uses a workload to illustrate it. kubernetes/examples is manifest-first: you get the YAML and a README explaining it. If you learn by reading explanations, the official tutorials are the better fit. If you learn by applying something and then reading the manifest to see what happened, this repository inverts the order usefully. The other alternative is Helm charts, whether from a project's own chart repository or a community index. The difference in approach is fundamental. A chart is parameterised and versioned, and installing it is a single command with values you override. An example here is a fixed set of manifests with a fixed configuration, and the point is that you read it. Charts optimise for repeatable deployment; this repository optimises for comprehension. Choosing a chart gets you something running faster. Choosing an example gets you an understanding of what running it requires, which is the thing you need before you can safely operate it. ## Maintenance, contribution and licence The repository is not archived, and its last push was on 2026-03-03. That is roughly six months before the date of writing, so the fair description is that it receives occasional updates rather than continuous development. There are no releases to track and therefore no upgrade path in the usual sense. Your cost of staying current is the cost of re-reading an example's README and re-checking its manifests against your cluster's API versions, per example, on your own schedule. Contributions are governed by two documents the README names explicitly: guidelines.md, which sets the standards, structure and best practices every example must follow, and CONTRIBUTING.md, which covers proposing, submitting and reviewing examples. If you plan to upstream an example, guidelines.md is the document that determines whether it is accepted, so read it before writing manifests rather than after. The licence is Apache-2.0. That is a permissive licence that generally allows reuse and modification with attribution and notice requirements, but the exact obligations depend on how you redistribute and on your jurisdiction. Read LICENSE in the repository for the terms that apply to you; this is a description of the licence, not legal advice. ## Conclusion Adopt kubernetes/examples if you learn Kubernetes by reading and applying manifests, or if you need a reference layout for AI, web, database and platform-app workloads. Do not adopt it if you want a supported distribution, a Helm chart catalogue or a versioned release you can pin; the repository has no releases, and the README does not document rollback. Before applying anything, open the README.md inside the specific subdirectory you care about, check its stated prerequisites (the top-level README says these include Kubernetes version and, where applicable, hardware), and confirm the manifest's API versions against your cluster. Start with one directory, not the whole tree. ## FAQ ### Do I need to install kubernetes/examples to use it? No. There is no package and no release; the README directs you to the subdirectories, and you work from a clone or from the files on the web. Each example's own README carries the deployment steps. ### Which Kubernetes version do the kubernetes/examples examples require? There is no single answer, because the top-level README states that each example's README covers its own prerequisites including Kubernetes version and hardware where applicable. Check the README inside the subdirectory you intend to use. ### Is kubernetes/examples still maintained? The repository is not archived, and its last push was on 2026-03-03. The top-level README states that examples should be kept up to date by the community and SIG Apps, and there is an _archived/ directory for examples that no longer meet the guidelines. ### What are the categories of examples in kubernetes/examples? The README names AI/ for AI/ML training, serving, platform reference manifests and MLOps toolchains, web/ for web applications from static sites to multi-tier apps, and databases/ for database deployment configurations. The repository also contains nginx-platform-app/ and an _archived/ directory. ## Sources - [Issues](https://github.com/kubernetes/examples/issues) - [kubernetes/examples on GitHub](https://github.com/kubernetes/examples) - [License: Apache-2.0](https://github.com/kubernetes/examples/blob/master/LICENSE) - [README](https://github.com/kubernetes/examples/blob/master/README.md) --- Hysen Labs editorial analysis, written from the project's own repository and release notes. Cite the canonical page: https://hysenlabs.com/projects/kubernetes-examples