kpt: Configuration as Data for Kubernetes
Shriira Press
A package-centric toolchain that treats Kubernetes configuration as data you can edit, transform, and ship — not templates you have to render.
Welcome to kpt: Configuration as Data for Kubernetes.
kpt is a CNCF Sandbox project that rethinks how teams manage Kubernetes configuration. Instead of hiding YAML behind templates or DSLs, kpt keeps the configuration itself as the source of truth — plain Kubernetes Resource Model (KRM) resources bundled into versioned packages in Git — and lets reusable, containerized functions transform and validate those resources in place. This book walks the whole story. It opens with the configuration problem kpt set out to solve and the "configuration as data" idea at its heart, then explains packages and the Kptfile, how packages are fetched and updated across upstream and downstream forks, and how KRM functions work both declaratively through the Kptfile pipeline and imperatively from the command line. From there it covers the function catalog, applying packages to live clusters with kpt live and ResourceGroup inventories, and Package Orchestration with Porch and Nephio. A closing chapter steps back to placement, comparisons with Helm and Kustomize, and practical adoption.
This title is part of the ShriIra library and is free to read in full, right here — our small contribution to making world-class knowledge easy to reach.
A note on reading it: open the Contents menu at the top of the reader to jump between chapters, use the Aa menu to set a comfortable text size, theme (light, sepia, or night), and single- or two-page layout. Your place is saved automatically, so you can always pick up where you left off.
We hope it serves you well.
— Shriira Press