kube-orchestrator 1.0.0

A full-featured Python SDK for Kubernetes

kube-orchestrator

A full-featured Python SDK for Kubernetes — programmatic kubectl, GitOps engine, platform engineering SDK.

---

Table of Contents

• Features
• Installation
• Quick Start
• Architecture
• Supported Resources
• Documentation
• Contributing
• License
• Acknowledgements

---

Features

• Full Kubernetes resource coverage — Pods, Deployments, StatefulSets, DaemonSets, Jobs, CronJobs, and more
• Programmatic kubectl — apply, delete, get, list, patch any resource from Python
• GitOps engine — load and apply YAML/JSON manifests with full validation
• Manifest Renderer — lightweight Helm-like templating engine
• Auto Rollback — detect failures and roll back deployments automatically
• Scaling Engine — HPA management and custom scaling watchers
• RBAC management — Roles, ClusterRoles, Bindings, ServiceAccounts
• Network management — Services, Ingress, NetworkPolicies, Endpoints
• Storage management — ConfigMaps, Secrets, PVs, PVCs, StorageClasses
• CRD Framework — install and manage Custom Resource Definitions
• CLI DevOps Tool — kube-orch command for terminal workflows
• Fully typed — PEP 561 compliant, works with mypy and pyright
• Python 3.10+ — modern Python, full type hints

---

Installation

pip install kube-orchestrator

---

Quick Start

from kube_orchestrator import KubeClient, PodManager

# Connect to cluster
client = KubeClient.from_kubeconfig()

# List all pods in a namespace
pod_manager = PodManager(client)
pods = pod_manager.list(namespace="default")
for pod in pods:
    print(pod.name, pod.status.phase)

from kube_orchestrator import KubeClient, DeploymentManager

client = KubeClient.from_kubeconfig()
deploy_manager = DeploymentManager(client)

# Scale a deployment
deploy_manager.scale(name="my-app", namespace="default", replicas=5)

from kube_orchestrator import KubeClient, ManifestApplier

client = KubeClient.from_kubeconfig()
applier = ManifestApplier(client)

# Apply a manifest file (like kubectl apply -f)
applier.apply_file("manifests/deployment.yaml")

from kube_orchestrator import KubeClient, AutoRollback

client = KubeClient.from_kubeconfig()
rollback = AutoRollback(client)

# Watch a deployment and auto-rollback on failure
rollback.watch(name="my-app", namespace="default", timeout=300)

from kube_orchestrator import KubeClient, ClusterHealthReporter

client = KubeClient.from_kubeconfig()
reporter = ClusterHealthReporter(client)

# Get a full cluster health report
report = reporter.full_report()
print(report.summary())

---

Architecture

kube-orchestrator
├── core/               # KubeClient, KubeConfig, logging, errors
├── resources/
│   ├── workloads/      # Pod, Deployment, StatefulSet, DaemonSet, Job, CronJob
│   ├── networking/     # Service, Ingress, NetworkPolicy, Endpoints
│   ├── storage/        # ConfigMap, Secret, PV, PVC, StorageClass
│   ├── rbac/           # Role, ClusterRole, Bindings, ServiceAccount
│   └── cluster/        # Namespace, Node, HPA, ResourceQuota, LimitRange
├── manifest/           # Loader, Validator, Renderer, Applier
├── crd/                # CRD installer and custom object manager
├── monitoring/         # ClusterHealthReporter
├── rollback/           # AutoRollback system
├── scaling/            # ScalingEngine, ResourceWatcher
└── cli/                # kube-orch CLI tool

---

Supported Resources

Resource	List	Get	Create	Update	Delete	Patch
Pod	✅	✅	✅	✅	✅	✅
Deployment	✅	✅	✅	✅	✅	✅
StatefulSet	✅	✅	✅	✅	✅	✅
DaemonSet	✅	✅	✅	✅	✅	✅
ReplicaSet	✅	✅	✅	✅	✅	✅
Job	✅	✅	✅	✅	✅	✅
CronJob	✅	✅	✅	✅	✅	✅
Service	✅	✅	✅	✅	✅	✅
Ingress	✅	✅	✅	✅	✅	✅
NetworkPolicy	✅	✅	✅	✅	✅	✅
ConfigMap	✅	✅	✅	✅	✅	✅
Secret	✅	✅	✅	✅	✅	✅
PersistentVolume	✅	✅	✅	✅	✅	✅
PersistentVolumeClaim	✅	✅	✅	✅	✅	✅
StorageClass	✅	✅	✅	✅	✅	✅
Namespace	✅	✅	✅	✅	✅	✅
Node	✅	✅	—	✅	—	✅
Role / ClusterRole	✅	✅	✅	✅	✅	✅
RoleBinding / ClusterRoleBinding	✅	✅	✅	✅	✅	✅
ServiceAccount	✅	✅	✅	✅	✅	✅
HPA	✅	✅	✅	✅	✅	✅
ResourceQuota	✅	✅	✅	✅	✅	✅
CRD	✅	✅	✅	✅	✅	✅

---

Documentation

Full documentation, including the API reference and guides, lives in the docs/ folder — start at docs/index.md.

---

Contributing

Contributions are welcome! Please read CONTRIBUTING.md before submitting a PR.

---

License

This project is licensed under the MIT License — see LICENSE for details.

---

Acknowledgements

• kubernetes-client/python — official Kubernetes Python client
• kubectl — inspiration for the CLI design