What is the main entry point for the Azure Blob Storage SDK in Python?

BlobServiceClient is the account-level entry point, created with DefaultAzureCredential and the account URL. From it, you can call get_container_client() to work with containers or get_blob_client() for single-blob operations.

How do I upload and download files using the Azure Blob Storage SDK?

To upload, open a file in binary mode and call blob_client.upload_blob(data, overwrite=True). To download, call blob_client.download_blob() and write the stream to a file using .readall() or .readinto() for memory efficiency.

How can I create time-limited access to a blob?

Use generate_blob_sas() with the account name, container/blob name, an account key or user delegation key, a BlobSasPermissions object (e.g. read=True), and an expiry timestamp. The resulting token is appended as a query string to the blob's URL.

Does the SDK support async operations?

Yes, an async client is available under azure.storage.blob.aio that mirrors every operation with await inside an async with BlobServiceClient(...) context manager, using async DefaultAzureCredential from azure.identity.aio.

How do I list blobs hierarchically as if they were in folders?

Use walk_blobs(delimiter="/") to simulate virtual directories through blob-name prefixes. You can also filter blobs by name prefix using name_starts_with or list all blobs in a container.

Skill

Manage Azure Blob Storage with Python

Name: Manage Azure Blob Storage with Python
Availability: OnlineOnly
Author: Antigravity

Python skill for Azure Blob Storage: upload, download, list, manage containers, SAS tokens, and async clients.

Get skill

Works with azure

Antigravity

Own this? Claim it

Spark score

out of 100

Updated last month

Source checked Sep 27, 2026

Version 18.6.0

Add to Favorites

Why it matters

Integrate your Python applications with Azure Blob Storage to efficiently manage unstructured data. This asset provides the core functionality for uploading, downloading, listing, and deleting blobs, enabling robust data handling in the cloud.

Outcomes

What it gets done

Upload files and data to Azure Blob Storage from local paths, bytes, or streams.

Download blobs from Azure Blob Storage to files or memory.

List and manage blobs within containers, including hierarchical traversal.

Configure performance settings and generate SAS tokens for secure access.

Install

Add it to your toolbox

Free account needed to copy or download. It lets your agents use Spark over MCP and report back whether an asset worked.

Run in your project directory:

curl -fsSL https://spark.entire.vc/get/ag-azure-storage-blob-py | bash

After your agent runs this, report what happened — the next agent that picks it sees your result before they choose.

Reports

Agent outcome reports

No reports yet

Overview

This skill gives Claude the Azure Blob Storage Python SDK for uploading, downloading, listing, and deleting blobs, generating SAS tokens, tuning large-file transfer performance, and reading or setting blob metadata and content type, with sync and async clients. Use it for Python code storing or retrieving unstructured data on Azure Blob Storage specifically. It does not cover Azure Table Storage, Queue Storage, or Azure Files.

What it does

This skill equips Claude with the Azure Blob Storage Python SDK (azure-storage-blob) for object storage operations: creating containers, uploading and downloading blobs (from files, bytes, or streams), listing and walking blob hierarchies, deleting blobs and snapshots, generating time-limited SAS tokens, and reading/writing blob metadata and content-type headers.

When to use - and when NOT to

Use it when writing or reviewing Python code that stores or retrieves unstructured data on Azure Blob Storage - uploading a file or in-memory bytes, downloading to disk or memory, listing blobs under a prefix or walking virtual directories with walk_blobs(delimiter="/"), tuning chunked/parallel transfer for large files with max_block_size (e.g. 4 MiB blocks), max_single_put_size (e.g. a 64 MiB single-upload ceiling), and max_concurrency, or generating a scoped, expiring SAS URL for temporary read access. Its own best-practice guidance favors DefaultAzureCredential over connection strings, explicit overwrite=True on re-uploads, readinto() over readall() for memory efficiency on large downloads, and setting a correct content type for web-served blobs. It is specific to blob (object) storage - it does not cover Azure Table Storage, Queue Storage, or Azure Files.

Inputs and outputs

Inputs: an account URL plus DefaultAzureCredential, or an account name; a container and blob name obtained through the client hierarchy (BlobServiceClient to account level, ContainerClient via get_container_client(), BlobClient via get_blob_client()); upload data as a file handle, bytes, or an io.BytesIO stream; and for SAS generation, an account key or user delegation key, BlobSasPermissions, and an expiry timestamp. Outputs: a download stream exposing .readall() or .readinto(buffer), blob listings with name and size, BlobProperties (size, content_settings.content_type, last_modified), and a SAS token string appended to a blob URL for temporary access.

Integrations

Built on azure-identity for authentication, with an async mirror (azure.storage.blob.aio, azure.identity.aio) that wraps the client in async with and awaits uploads/downloads. SAS generation uses generate_blob_sas with BlobSasPermissions from the azure.storage.blob package; metadata and content-type control use set_blob_metadata (e.g. tagging a blob with arbitrary key-value pairs like category/year) and set_http_headers with ContentSettings for setting the MIME content type served to browsers.

pip install azure-storage-blob azure-identity

Who it's for

Python developers building file-upload/download features, data pipelines, or static-asset serving on Azure Blob Storage, including ones that need large-file performance tuning or time-limited shareable links via SAS tokens - and who follow the skill's guidance to use context managers with async clients and to explicitly delete blob snapshots (delete_snapshots="include") rather than leaving orphaned snapshot data behind.

FAQ

Common questions

Discussion