azure-containerregistry-py

作者: microsoft

Azure Container Registry SDK for Python. Use for managing container images, artifacts, and repositories. Triggers: "azure-containerregistry", "ContainerRegistryClient", "container images", "docker registry", "ACR".

npx skills add https://github.com/microsoft/skills --skill azure-containerregistry-py

Azure Container Registry SDK for Python

Manage container images, artifacts, and repositories in Azure Container Registry.

Installation

pip install azure-containerregistry

Environment Variables

AZURE_CONTAINERREGISTRY_ENDPOINT=https://<registry-name>.azurecr.io  # Required for all auth methods
AZURE_TOKEN_CREDENTIALS=prod # Required only if DefaultAzureCredential is used in production

Authentication & Lifecycle

🔑 Two rules apply to every code sample below:

  1. Prefer DefaultAzureCredential. It works locally (Azure CLI / VS Code / Developer CLI) and in Azure (managed identity, workload identity) with no code change. Avoid connection strings, account/API keys — they bypass Entra audit and rotation.
    • Local dev: DefaultAzureCredential works as-is.
    • Production: set AZURE_TOKEN_CREDENTIALS=prod (or AZURE_TOKEN_CREDENTIALS=<specific_credential>) to constrain the credential chain to production-safe credentials.
  2. Wrap every client in a context manager so HTTP transports, sockets, and token caches are released deterministically:
    • Sync: with <Client>(...) as client:
    • Async: async with <Client>(...) as client: and async with DefaultAzureCredential() as credential: (from azure.identity.aio)

Snippets may abbreviate this setup, but production code should always follow both rules.

Entra ID (Recommended)

import os
from azure.containerregistry import ContainerRegistryClient
from azure.identity import DefaultAzureCredential, ManagedIdentityCredential

# Local dev: DefaultAzureCredential. Production: set AZURE_TOKEN_CREDENTIALS=prod or AZURE_TOKEN_CREDENTIALS=<specific_credential>
credential = DefaultAzureCredential(require_envvar=True)
# Or use a specific credential directly in production:
# See https://learn.microsoft.com/python/api/overview/azure/identity-readme?view=azure-python#credential-classes
# credential = ManagedIdentityCredential()

with ContainerRegistryClient(
    endpoint=os.environ["AZURE_CONTAINERREGISTRY_ENDPOINT"],
    credential=credential
) as client:
    # Use client here (see following sections for operations)
    ...

Anonymous Access (Public Registry)

from azure.containerregistry import ContainerRegistryClient

with ContainerRegistryClient(
    endpoint="https://mcr.microsoft.com",
    credential=None,
    audience="https://mcr.microsoft.com"
) as client:
    # Use client here (see following sections for operations)
    ...

List Repositories

with ContainerRegistryClient(endpoint, DefaultAzureCredential()) as client:
    for repository in client.list_repository_names():
        print(repository)

Repository Operations

Get Repository Properties

properties = client.get_repository_properties("my-image")
print(f"Created: {properties.created_on}")
print(f"Modified: {properties.last_updated_on}")
print(f"Manifests: {properties.manifest_count}")
print(f"Tags: {properties.tag_count}")

Update Repository Properties

from azure.containerregistry import RepositoryProperties

client.update_repository_properties(
    "my-image",
    properties=RepositoryProperties(
        can_delete=False,
        can_write=False
    )
)

Delete Repository

client.delete_repository("my-image")

List Tags

for tag in client.list_tag_properties("my-image"):
    print(f"{tag.name}: {tag.created_on}")

Filter by Order

from azure.containerregistry import ArtifactTagOrder

# Most recent first
for tag in client.list_tag_properties(
    "my-image",
    order_by=ArtifactTagOrder.LAST_UPDATED_ON_DESCENDING
):
    print(f"{tag.name}: {tag.last_updated_on}")

Manifest Operations

List Manifests

from azure.containerregistry import ArtifactManifestOrder

for manifest in client.list_manifest_properties(
    "my-image",
    order_by=ArtifactManifestOrder.LAST_UPDATED_ON_DESCENDING
):
    print(f"Digest: {manifest.digest}")
    print(f"Tags: {manifest.tags}")
    print(f"Size: {manifest.size_in_bytes}")

Get Manifest Properties

manifest = client.get_manifest_properties("my-image", "latest")
print(f"Digest: {manifest.digest}")
print(f"Architecture: {manifest.architecture}")
print(f"OS: {manifest.operating_system}")

Update Manifest Properties

from azure.containerregistry import ArtifactManifestProperties

client.update_manifest_properties(
    "my-image",
    "latest",
    properties=ArtifactManifestProperties(
        can_delete=False,
        can_write=False
    )
)

Delete Manifest

# Delete by digest
client.delete_manifest("my-image", "sha256:abc123...")

# Delete by tag
manifest = client.get_manifest_properties("my-image", "old-tag")
client.delete_manifest("my-image", manifest.digest)

Tag Operations

Get Tag Properties

tag = client.get_tag_properties("my-image", "latest")
print(f"Digest: {tag.digest}")
print(f"Created: {tag.created_on}")

Delete Tag

client.delete_tag("my-image", "old-tag")

Upload and Download Artifacts

from azure.containerregistry import ContainerRegistryClient

with ContainerRegistryClient(endpoint, DefaultAzureCredential()) as client:
    # Download manifest
    manifest = client.download_manifest("my-image", "latest")
    print(f"Media type: {manifest.media_type}")
    print(f"Digest: {manifest.digest}")

    # Download blob
    blob = client.download_blob("my-image", "sha256:abc123...")
    with open("layer.tar.gz", "wb") as f:
        for chunk in blob:
            f.write(chunk)

Async Client

from azure.containerregistry.aio import ContainerRegistryClient
from azure.identity.aio import DefaultAzureCredential

async def list_repos():
    async with DefaultAzureCredential() as credential:
        async with ContainerRegistryClient(endpoint, credential) as client:
            async for repo in client.list_repository_names():
                print(repo)

Clean Up Old Images

from datetime import datetime, timedelta, timezone

cutoff = datetime.now(timezone.utc) - timedelta(days=30)

for manifest in client.list_manifest_properties("my-image"):
    if manifest.last_updated_on < cutoff and not manifest.tags:
        print(f"Deleting {manifest.digest}")
        client.delete_manifest("my-image", manifest.digest)

Client Operations

OperationDescription
list_repository_namesList all repositories
get_repository_propertiesGet repository metadata
delete_repositoryDelete repository and all images
list_tag_propertiesList tags in repository
get_tag_propertiesGet tag metadata
delete_tagDelete specific tag
list_manifest_propertiesList manifests in repository
get_manifest_propertiesGet manifest metadata
delete_manifestDelete manifest by digest
download_manifestDownload manifest content
download_blobDownload layer blob

Best Practices

  1. Pick sync OR async and stay consistent. Do not mix azure.xxx sync clients with azure.xxx.aio async clients in the same call path. Choose one mode per module.
  2. Always use context managers for clients and async credentials. Wrap every client in with Client(...) as client: (sync) or async with Client(...) as client: (async). For async DefaultAzureCredential from azure.identity.aio, also use async with credential: so tokens and transports are cleaned up.
  3. Use Microsoft Entra ID for authentication in production
  4. Delete by digest not tag to avoid orphaned images
  5. Lock production images with can_delete=False
  6. Clean up untagged manifests regularly
  7. Use async client for high-throughput operations
  8. Order by last_updated to find recent/old images
  9. Check manifest.tags before deleting to avoid removing tagged images

Reference Files

FileContents
references/capabilities.mdAdditional non-hero capabilities, operation-group coverage, and production checklists.
references/non-hero-scenarios.mdDedicated non-hero examples for secondary/advanced scenarios.

来自 microsoft 的更多技能

oss-growth
microsoft
OSS增长黑客角色
agent-framework-azure-ai-py
microsoft
使用Microsoft Agent Framework Python SDK(agent-framework-azure-ai)构建Azure AI Foundry代理。在创建使用AzureAIAgentsProvider的持久化代理、使用托管工具(代码解释器、文件搜索、网络搜索)、集成MCP服务器、管理对话线程或实现流式响应时使用。涵盖函数工具、结构化输出和多工具代理。
development
airunway-aks-setup
microsoft
Set up AI Runway on AKS — from bare cluster to running model. Covers cluster verification, controller install, GPU assessment, provider setup, and first deployment. WHEN: "setup AI Runway", "onboard AKS cluster", "install AI Runway", "airunway setup", "deploy model to AKS", "GPU inference on AKS", "KAITO setup on AKS", "run LLM on AKS", "vLLM on AKS", "set up model serving on AKS", "AI Runway controller".
devops
appinsights-instrumentation
microsoft
使用Azure Application Insights对Web应用进行插桩的指南。提供遥测模式、SDK设置和配置参考。适用场景:如何对应用进行插桩、App Insights SDK、遥测模式、什么是App Insights、Application Insights指南、插桩示例、APM最佳实践。
devops
applicationinsights-web-ts
microsoft
使用Application Insights JavaScript SDK(@microsoft/applicationinsights-web)为浏览器/Web应用添加检测。用于真实用户监控(RUM)——页面视图、点击、AJAX/fetch依赖项、异常、自定义事件,以及与后端OpenTelemetry追踪关联的浏览器端GenAI代理追踪。涵盖SDK加载器脚本和npm设置、框架扩展(React、React Native、Angular)、点击分析、遥测初始化器,以及从浏览器发出的代理/工具/模型跨度所遵循的OTel GenAI语义约定。
devops
azure-ai-anomalydetector-java
microsoft
使用适用于 Java 的 Azure AI 异常检测器 SDK 构建异常检测应用程序。在实现单变量/多变量异常检测、时间序列分析或 AI 驱动的监控时使用。
development
azure-ai-language-conversations-py
microsoft
使用azure-ai-language-conversations Python SDK实现对话语言理解(CLU)。当使用ConversationAnalysisClient分析对话意图和实体、构建NLP功能或将语言理解集成到应用程序中时使用。
development
azure-ai-ml-py
microsoft
Azure Machine Learning SDK v2 for Python。用于机器学习工作区、作业、模型、数据集、计算资源和管道。 触发词:“azure-ai-ml”、“MLClient”、“工作区”、“模型注册表”、“训练作业”、“数据集”。
development