---
id: kfp-kubernetes
version: "2.17.0"
license: Apache 2.0
license_treatment: permissive
maintenance: active
---

# kfp-kubernetes — Kubernetes platform configuration library and generated protos.

License: permissive · Maintenance: active · Downloads: 265.7K/mo

## What it is and what it does

kfp-kubernetes is an addon library to Kubeflow Pipelines SDK that brings native Kubernetes concepts into pipeline authoring. Instead of writing generic pipeline code, you can directly configure Kubernetes-level details like Secrets, PersistentVolumeClaims, node selectors, tolerations, image pull policies, and pod metadata directly on pipeline tasks. This bridges the gap between high-level pipeline logic and low-level Kubernetes pod configuration.

The library works by providing a set of functions that modify task definitions to inject Kubernetes-specific configuration. You define your pipeline components normally using the kfp SDK, then call functions like `use_secret_as_env`, `mount_pvc`, `add_pod_label`, or `set_timeout` to layer on Kubernetes behavior. This approach keeps pipeline logic separate from infrastructure concerns while giving you full control over how tasks run on the cluster.

Use it for:

- Mount Kubernetes Secrets as environment variables or volumes in pipeline task containers for credential and configuration management.
- Attach PersistentVolumeClaims to tasks for shared data storage across pipeline stages or ephemeral volumes for temporary task-local storage.
- Configure pod scheduling constraints using node selectors and tolerations to run tasks on specific cluster nodes or hardware.
- Set pod metadata like labels and annotations for observability, cost tracking, or integration with external monitoring systems.
- Control container image pull behavior and set task timeouts using Kubernetes pod spec fields directly from pipeline code.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Extends Kubeflow Pipelines SDK with Kubernetes-native features like Secrets, PersistentVolumeClaims, node selectors, tolerations, and pod metadata configuration for pipeline tasks.

Yes, if you are authoring Kubeflow Pipelines and need to configure Kubernetes-specific features like storage, secrets, or scheduling. The package is actively maintained, has low install friction, carries a permissive license, and integrates cleanly with the kfp SDK. Install it only if you are already using Kubeflow Pipelines; it is not useful standalone.

## Install

pip install kfp-kubernetes
uv add kfp-kubernetes
poetry add kfp-kubernetes

## Installing kfp-kubernetes

Before you install: Low install friction with a pure-Python wheel distribution. Active maintenance with recent releases; last commit 2026-08-14 and latest release 2026-07-09. Requires Python >=3.9.0.

License in practice: Apache 2.0 permissive license allows commercial and private use with minimal restrictions, typical for open-source infrastructure tools.

Quickstart:

pip install kfp-kubernetes

from kfp import dsl
from kfp import kubernetes

@dsl.component
def my_task():
    print('hello')

@dsl.pipeline
def my_pipeline():
    task = my_task()
    kubernetes.use_secret_as_env(task, secret_name='my-secret', secret_key_to_env={'password': 'SECRET_VAR'})

Requires an active Kubeflow Pipelines installation and a Kubernetes cluster to run pipelines; Python >=3.9.0 required.

Verify before relying:

- Whether this package works standalone or requires a specific version of kfp to be installed first.
- Compatibility with different Kubernetes versions or Kubeflow Pipelines versions beyond the runtime dependency on kfp.
- Performance characteristics when managing large numbers of volumes or secrets across pipeline tasks.

## Package facts

- License: Apache 2.0 (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 265.7K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

kubeflow pipelines kubernetes integration, kfp kubernetes secrets volumes, kubeflow pipeline pod configuration, kubernetes persistent volumes kfp, kubeflow pipeline node selectors, kfp kubernetes manifests, kubeflow pipeline storage claims, kubeflow, kubernetes-integration, ml-pipelines

[View on SkillFed](https://skillfed.io/packages/kfp-kubernetes) · [View on PyPI](https://pypi.org/project/kfp-kubernetes/)