Azure Blob Storage

Manage Azure Blob Storage containers and blobs from FlowRunner agents: list, create, upload, download, copy, snapshot, and manage metadata, with hand-rolled Shared Key request signing.

13 actions available

A flow finishes processing a document and needs durable storage

Agent reads the file it produced or received into the flow

Agent uploads the blob to a container from inline content or a source URL

Agent sets object metadata tying the blob to the source record

Agent confirms the upload by reading the blob properties

Agent posts the container path and metadata to the processing channel

A person confirms any container deletion before it removes every blob inside

What This Integration Enables

The Azure Blob Storage connector covers containers and blobs end to end. Agents list containers with prefix filtering and paging, create containers with an access level and metadata, read container properties, and delete containers. On blobs, agents list with a prefix, upload from inline text or a streamed source URL, download into FlowRunner file storage, and read blob properties without downloading. They copy blobs server-side from another Azure blob or accessible URL, snapshot a blob to preserve a point-in-time version, and set or read custom metadata as plain name/value pairs. Because the bytes for a copy never route through the flow, copies are fast at any size. Some of these actions read and some destroy; the connector supplies both, and the surrounding flow decides what needs a person.

Without FlowRunner

Documents filed by hand Files land in inconsistent folders with no tie to the record they belong to

Retrieval is a search Finding an old file means checking several locations with no reliable key

Deletes run unguarded A cleanup step can remove a container and everything in it with no second look

With FlowRunner

Documents filed on upload Every blob is written with metadata tied to the source record automatically

Retrieval by key Any stored file is fetched by its container path and metadata in seconds

Deletes gated Container deletion pauses for confirmation before it removes every blob inside

Use Case Scenarios

Document Archive With Metadata

An agent finishes validating a vendor invoice and entering it into an ERP. It uploads the original PDF to a container with Upload Blob, then calls Set Blob Metadata to attach the vendor name, amount, and ERP record ID. Months later, when an auditor asks for that invoice, the agent lists the container by prefix and pulls the exact blob by its metadata. No manual filing, and no lost documents.

Snapshot Before a Change

Before a flow overwrites a stored config file or a working document, the agent calls Snapshot Blob to preserve the current version by timestamp. If the new write turns out wrong, the prior snapshot is still there to restore. The point-in-time copy costs almost nothing because a snapshot shares storage with the base blob until they diverge.

Mirror Incoming Files

When a file arrives in another storage service, the agent calls Upload Blob with the file's source URL to mirror it into an Azure container for long-term retention. The file streams straight into the blob without routing its bytes through the flow, so even large files land quickly in the archive.

Human-in-Loop Highlight

Listing containers, reading properties, and uploading blobs are safe operations a flow runs on its own. Deleting a container is not: the call marks the container and every blob it holds for deletion, and the operation cannot be undone. This is [human-in-the-loop](/concepts/human-in-the-loop/) at the boundary between routine storage work and destroying data at scale. Before deleting a container, the agent pauses, assembles the container name and a count of the blobs inside, and routes to a person through their preferred channel. The message is specific: "This would delete the container invoices-2024 and all 3,180 blobs it holds. This cannot be undone. Confirm the deletion, or cancel?" The agent handles the archiving, the metadata, and the retrieval; a person owns the call that erases a container's contents.

Agent processes routinely

Detects exception requiring judgment

Clear match Continues automatically

Ambiguous Routes to human via preferred channel

Human decides

Agent resumes with decision

Agent Capabilities

13 actions

Containers

List Containers Lists the blob containers in the storage account, with optional name-prefix filtering and pagination. Use it to drive dynamic flows that iterate over containers.
Create Container Creates a new container under the given name, with an optional public access level and custom metadata. Use it when provisioning storage for a new project or client.
Get Container Properties Retrieves system properties and custom metadata for a container, including ETag, last-modified time, and access level. Use it to confirm a container exists before writing to it.
Delete Container Marks a container and every blob it holds for deletion. The operation cannot be undone, so it routes through human confirmation for shared containers.

Blobs

List Blobs Lists the blobs inside a container, filtered by name prefix and paged. Use it in batch flows that process every file under a path.
Upload Blob Uploads a block blob from inline text or a source URL, streaming the file at that URL into the blob. Use it to archive documents produced or received by a flow.
Get Blob Downloads a blob into FlowRunner file storage and returns a URL that downstream steps can attach, forward, or process. Use it to bring an Azure file into a flow.
Get Blob Properties Retrieves system properties and metadata for a blob without downloading its contents. Use it to inspect size and content type before deciding whether to download.
Delete Blob Deletes a blob from a container. When soft-delete is enabled the blob is retained for the configured window before being purged.
Copy Blob Copies a blob server-side from another Azure blob or accessible URL. The copy runs asynchronously and is fast because the bytes never route through the flow.
Snapshot Blob Creates a read-only, point-in-time snapshot of a blob. Use it to preserve a version before a flow modifies or deletes the base blob.

Metadata

Set Blob Metadata Replaces all custom metadata on a blob with the supplied name/value pairs. Use it to tag objects with the record IDs a flow needs for later retrieval.
Get Blob Metadata Retrieves only the custom metadata on a blob, without its contents or full properties. Use it for a lightweight read of user-defined tags.

Start building with Azure Blob Storage

$100 in credits. No card required. Connect in minutes.

Start Building Free Talk to us