Python bindings for the Helm v4 Go library. Ships with a self-contained Go shim that bundles the native Helm runtime—no system dependencies required.

✨ Features

• 🚀 Complete Helm API - Full access to Helm v4 functionality
• ⚡ Async-First - All operations use async/await for non-blocking execution
• 📦 Self-Contained - No system dependencies required
• 🔒 Type-Safe - Python type hints for better IDE support
• 🐍 Pythonic API - Intuitive, object-oriented interface
• 🎯 Production Ready - Based on official Helm Go libraries
• 🔄 Concurrent - Execute multiple operations simultaneously with asyncio

🚀 Quick Start

Note: helm-sdkpy uses an async-first API. All operations must be awaited. This enables non-blocking execution and concurrent operations with asyncio.gather().

Installation

pip install helm-sdkpy

Basic Usage

import asyncio
import helm-sdkpy

async def main():
    # Create a configuration
    config = helm-sdkpy.Configuration(namespace="default")

    # Install a chart from a local path
    install = helm-sdkpy.Install(config)
    result = await install.run(
        release_name="my-nginx",
        chart_path="./nginx-chart",
        values={"replicaCount": 3}
    )
    print(f"Installed: {result['name']}")

    # Install a chart from an OCI registry
    result = await install.run(
        release_name="my-app",
        chart_path="oci://ghcr.io/nginxinc/charts/nginx-ingress",
        values={"controller": {"service": {"type": "LoadBalancer"}}}
    )
    print(f"Installed: {result['name']}")

    # Install a chart from an HTTPS URL
    result = await install.run(
        release_name="my-release",
        chart_path="https://charts.bitnami.com/bitnami/nginx-15.0.0.tgz",
        values={"replicaCount": 2}
    )
    print(f"Installed: {result['name']}")

    # List releases
    list_action = helm-sdkpy.List(config)
    releases = await list_action.run(all=True)
    for release in releases:
        print(f"Release: {release['name']}")

    # Upgrade a release
    upgrade = helm-sdkpy.Upgrade(config)
    result = await upgrade.run(
        release_name="my-nginx",
        chart_path="./nginx-chart",
        values={"replicaCount": 5}
    )

    # Uninstall a release
    uninstall = helm-sdkpy.Uninstall(config)
    result = await uninstall.run("my-nginx")

asyncio.run(main())

Concurrent Operations

Take advantage of Python's asyncio to run multiple Helm operations concurrently:

import asyncio
import helm-sdkpy

async def deploy_multiple_apps():
    config = helm-sdkpy.Configuration(namespace="default")
    install = helm-sdkpy.Install(config)
    
    # Install multiple charts concurrently
    results = await asyncio.gather(
        install.run("app-1", "oci://registry.io/chart1"),
        install.run("app-2", "oci://registry.io/chart2"),
        install.run("app-3", "oci://registry.io/chart3"),
    )
    
    for result in results:
        print(f"Deployed: {result['name']}")

asyncio.run(deploy_multiple_apps())

📖 Chart Path Formats

helm-sdkpy supports multiple chart location formats:

Local Paths

Point to a chart directory or packaged chart (.tgz) on your local filesystem:

chart_path="./nginx-chart"           # Local directory
chart_path="/path/to/mychart"        # Absolute path
chart_path="./mychart-1.0.0.tgz"     # Packaged chart

OCI Registries

Reference charts stored in OCI-compatible container registries:

chart_path="oci://ghcr.io/nginxinc/charts/nginx-ingress"
chart_path="oci://registry.example.com/charts/myapp"

HTTP/HTTPS URLs

Download charts directly from web servers:

chart_path="https://charts.bitnami.com/bitnami/nginx-15.0.0.tgz"
chart_path="https://example.com/charts/myapp-1.2.3.tgz"

🔧 Kubeconfig Configuration

helm-sdkpy provides flexible options for configuring Kubernetes cluster access:

Default Kubeconfig

# Uses $KUBECONFIG env var or ~/.kube/config
config = helm_sdkpy.Configuration(namespace="default")

File Path

# Explicit path to kubeconfig file
config = helm_sdkpy.Configuration(
    namespace="default",
    kubeconfig="/path/to/kubeconfig.yaml"
)

YAML String

Pass kubeconfig content directly as a string - useful for dynamic configurations, secrets, or CI/CD environments:

# Kubeconfig from environment variable
kubeconfig_content = os.environ.get("KUBECONFIG_CONTENT")

# Or read from a secret, API response, etc.
config = helm_sdkpy.Configuration(
    namespace="default",
    kubeconfig=kubeconfig_content  # YAML string auto-detected
)

Specific Context

# Use a specific context from multi-cluster kubeconfig
config = helm_sdkpy.Configuration(
    namespace="production",
    kubeconfig="/path/to/kubeconfig.yaml",
    kubecontext="production-cluster"
)

See examples/kubeconfig_usage.py for more detailed examples.

📖 API Overview

Core Actions (All Async)

• Configuration - Manage Helm configuration and Kubernetes connection
• Install - Install charts to Kubernetes (async)
• Upgrade - Upgrade existing releases (async)
• Uninstall - Remove releases from cluster (async)
• List - List deployed releases (async)
• Status - Get release status information (async)
• Rollback - Rollback to previous release versions (async)
• GetValues - Retrieve release values (async)
• History - View release history (async)

Chart Operations (All Async)

• Pull - Download charts from repositories (async)
• Show - Display chart information and values (async)
• Test - Run release tests (async)
• Lint - Validate charts for errors (async)
• Package - Package charts into archives (async)

All methods use async def and must be awaited. This enables non-blocking operations and concurrent execution with asyncio.gather().

🛠️ Development

Prerequisites

• Python 3.12+
• Docker (for building the native library)
• just task runner (optional but recommended)

Building from Source

git clone https://github.com/vantagecompute/helm-sdkpy.git
cd helm-sdkpy

# Build native library using Docker
just build-lib

# Install in development mode
pip install -e .

Running Tests

# Run test suite
just unit

# Run linting
just lint

# Run type checking
just typecheck

Project Commands

The project uses just for task automation:

# Install just
sudo snap install just --classic

Available commands:

• just build-lib - Build native library (Docker)
• just unit - Run tests with coverage
• just lint - Check code style
• just typecheck - Run static type checking
• just fmt - Format code

📝 Examples

See the examples/ directory for more usage examples.

🏗️ Architecture

helm-sdkpy follows the same architecture as dqlitepy:

1. Go Shim Layer - Exposes Helm v4 API via C FFI
2. Python FFI - CFFI bindings to Go shared library
3. Python API - Pythonic wrapper classes

This ensures type safety, excellent performance, and seamless integration with the official Helm libraries.

📝 License

This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

Copyright 2025 Vantage Compute

🤝 Contributing

Contributions welcome! Please ensure:

• Code follows existing style (ruff formatting)
• Tests pass and coverage is maintained
• Type hints are included
• Documentation is updated

💬 Support

• GitHub Issues - Bug reports and feature requests
• Examples - Sample code and use cases