SellIntegro CloudPrint

SellIntegro CloudPrint sends documents from a flow to a physical printer. Agents print a shipping label on a warehouse thermal printer the moment an order is ready, or an invoice or packing slip when it is paid, from a PDF or raw ZPL and EPL commands.

7 actions API key available

SellIntegro CloudPrint website Platform Documentation Capability data verified 2026-08-26

An order reaches ready to ship and its courier label has already been bought upstream

Agent pulls the order and the label document produced earlier in the flow

List Printers confirms the target printer's host computer is connected, not merely that the printer exists

Print PDF uploads the label bytes to the bench printer

Get Print Job polls the job until its state moves from queued to printed

The packing bench channel gets the order number and the printer it went to

A packer checks the first label of a run against the order before the rest of the queue is released

What This Integration Enables

Almost every step in a workflow can be taken back. A record can be updated, a message can be edited, a draft can be deleted. CloudPrint is where that stops being true. It takes a document from a flow and puts it on a physical printer, and once the paper is out there is no cancel operation, no delete operation, and nothing in this connector that can reach back into the queue and stop a job. That is not a shortcoming to work around. It is the reason this connector belongs at the very end of a flow, after everything that could still be wrong has already been settled.

FlowRunner agents use it to close the last gap in a fulfillment process: the label a warehouse thermal printer produces the moment an order is picked, the invoice that prints when a payment clears, the packing slip that comes out beside them. Documents can go as a PDF uploaded as bytes, as a PDF that CloudPrint fetches from a URL itself, or as raw ZPL or EPL commands sent straight to a label printer with no rendering step in between. The connector reads printer state before it commits, and it polls a job to completion afterwards, so the flow knows what happened rather than assuming.

Without FlowRunner

A person between the flow and the printer Documents are generated automatically and then downloaded and printed by hand

Printer state discovered at the bench Nobody knows the warehouse machine is offline until a run produces nothing

Labels printed from the wrong screen The packer prints from the courier portal, so the order and the label are matched by eye

With FlowRunner

The flow reaches the hardware The workflow that decides a document is ready is the workflow that sends it to a named printer

Availability checked before committing The host computer's connection state is read before a run is allowed to start

Order and document stay bound The label that prints is the one the flow attached to that order, and the job id proves it

Use Case Scenarios

Label at the bench the moment the order is ready

An order in Shopify moves to ready to ship. The agent reads the line items, buys a label through Shippo, and holds the label URL. Before it prints anything it calls List Printers and checks the nested computer's connection state for the bench machine, because CloudPrint accepts a job for a printer whose host is offline and simply leaves it sitting in the queue. With the host confirmed connected, Print PDF sends the label bytes, and Get Print Job polls until the state reads printed. The packer at the bench gets a label without opening a courier portal.
Courier labels from an existing fulfillment stack

A warehouse already runs its couriers through BaseLinker. The agent fetches the courier label from there and passes it to CloudPrint by URL rather than by upload, which removes the size ceiling entirely because CloudPrint fetches the document server side. The tradeoff is stated plainly in the connector: that URL has to be reachable from SellIntegro's own infrastructure, so a signed or expiring link may fail after the job was already accepted. For anything comfortably inside the upload limit, sending the bytes is the more predictable path.
Documents that follow the money

A payment clears in Stripe and the invoice is raised in the accounting system. The agent renders the invoice and the packing slip, prints them to the office printer, and sends only the label to the warehouse thermal printer, resolving both by id from a single List Printers call. Each print operation returns a job id, and the flow stores it against the order, because the queue listing itself is capped and cannot be paged, so a job id kept at the time of printing is the only reliable way to read that job back later.

Human-in-Loop Highlight

Print Raw Label passes ZPL or EPL commands to a thermal printer untouched, with no rendering and no validation. A dialect the printer does not speak does not produce an API error. It produces a blank or garbled label, and CloudPrint reports the job as printed, because from its side it was. Combine that with the fact that there is no cancel operation anywhere in this connector, and a bulk label run is a decision that cannot be walked back once it starts. So the flow prints one. The agent sends the first label of the run, waits for Get Print Job to report it printed, and asks the packing bench: "First label of 240 for the Tuesday dispatch is out on Printer 3. Scan it to confirm the barcode reads and the address block is complete, then approve the remaining 239." The packer scans it. Only then does the rest of the queue go. The agent is faster than a person at everything in this flow except knowing what a real label looks like coming off that particular machine.

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

7 actions

Account

Get Account Returns the CloudPrint account the API key belongs to, with its identifier, email, and state. This is the connection test, and it is the only route that answers a truthful authentication failure for a bad key, because CloudPrint validates a print job's body before it authenticates the caller.

Printers

List Printers Returns every printer this account can print to, each with its identifier, name, description, state, default flag, capabilities, and the computer that hosts it. A printer is only usable while its host computer is connected: a job sent to one whose computer is disconnected is accepted and then sits in the queue, so check that field before printing. This is also the only way to learn a printer identifier.

Print Jobs

Print PDF Sends a PDF to a printer by uploading its bytes, which is the form to prefer for anything CloudPrint cannot fetch itself. CloudPrint caps the encoded content, and this operation refuses an oversized document with a clear message rather than letting the API answer a success that hides a rejection. There is no way to cancel the job once it is queued.
Print PDF From URL Sends a PDF to a printer by URL, which CloudPrint fetches server side, so this form has no size ceiling. The URL must be publicly reachable from SellIntegro's infrastructure: a signed, expiring, or private link, including a FlowRunner file storage URL, may not resolve for them, and the job then fails after it was already accepted. Prefer uploading the bytes for anything inside the size cap.
Print Raw Label Sends raw printer command language such as ZPL or EPL straight to a thermal label printer, bypassing any document rendering. The commands stay under CloudPrint's encoded size cap and are passed through untouched, so a dialect the printer does not speak produces a blank or garbled label rather than an API error.
List Print Jobs Returns the print queue, which the vendor defines as the most recent entries or everything from the first waiting job onward. This is not a complete history and it cannot be paged, because CloudPrint publishes no page, offset, limit, or cursor parameter, so a busy account cannot enumerate its past jobs. Keep the identifier returned by a print operation and read that job back instead.
Get Print Job Returns a single print job by identifier, which is how a flow polls a document to completion by watching its state move from queued to printed. This route is undocumented: SellIntegro's API reference does not mention it, and it was confirmed to exist by probing, so its response shape is inferred from the shape of an entry in the queue listing. There is no cancel or delete counterpart.

Frequently Asked Questions

What can FlowRunner do with SellIntegro CloudPrint?

FlowRunner agents can run Get Account, List Printers, and Print PDF in SellIntegro CloudPrint, plus 4 more actions.

Does connecting SellIntegro CloudPrint to FlowRunner require OAuth?

No. SellIntegro CloudPrint connects to FlowRunner with an API key, no OAuth flow required.

Can SellIntegro CloudPrint trigger a FlowRunner workflow automatically?

SellIntegro CloudPrint doesn't currently expose triggers in FlowRunner. It connects as an action step inside workflows started by another trigger.

Start building with SellIntegro CloudPrint

Free plan, no card required. Connect in minutes.

Start Building Free Talk to us