Dynamic Mockups

Dynamic Mockups renders product mockups by placing artwork into PSD smart objects. Agents render single mockups, batches, or whole collections, export print files, generate AI mockup scenes, and animate a still into a short motion video.

Verified 23 actions API key available

Dynamic Mockups website Platform Documentation Capability data verified 2026-08-28

A new artwork is approved for the seasonal drop

Agent reads the artwork file, the colorway list, and the templates the drop covers

Get Mockup returns the smart objects each template offers, since there is no way to guess their UUIDs

Render Collection places the one design across every template in the collection

Agent reads failed_renders rather than trusting the success flag, and requeues the misses

The rendered set posts to the merchandising channel with the failures called out

Production approves the scope before Export Print Files spends per file and reaches the supplier

What This Integration Enables

Dynamic Mockups renders two different things from the same request, and only one of them is a picture. The mockup image is what a shopper sees on a listing. The print file is what a supplier feeds into a press or an embroidery machine, and it becomes a physical object in a box. Treating those as the same operation is how a product ships with artwork positioned two centimeters off, five hundred times. This connector keeps them distinct: Render Mockup and Render Collection produce the listing images, Export Print Files produces the production artwork from the identical placement, and it bills per print file rather than per call.

FlowRunner agents work the whole path. They read a template to learn which smart objects it offers, because a render is addressed by mockup and smart object UUIDs and there is no way to guess them. They place artwork and colors across a collection in one call, generate scenes with the AI surface when no PSD exists, ground those scenes on a real print-on-demand garment rather than an invented one, and animate a still into a short video for a paid social slot. What they do not do is decide how much of that reaches a supplier, because that decision spends credits and produces goods. That boundary is where FlowRunner's human-in-the-loop design does its work.

Without FlowRunner

Listing images gate the launch Every colorway and every product needs its own mockup, so the drop ships when the design queue clears

The catalog goes stale by attrition A new garment or a new template means redoing artwork placement by hand, so most listings keep last season's images

Print files assembled by hand Someone exports the production file per product, and the wrong DPI or the wrong decoration area is found by the supplier

With FlowRunner

Images produced from the artwork One approved design renders across every template in a collection in a single call

The catalog keeps up with the templates Adding a template to a collection is enough to bring every existing design onto it

Print files produced from the same request The production file comes from the same placement the listing image used, at the DPI the supplier asked for

Use Case Scenarios

A drop that lists itself

An artwork clears review and the agent picks it up. Get Mockup reads each template's smart objects so the artwork, the colorway hex values, and the fit mode land on the right layers. Render Collection then applies that one design across every template in the seasonal collection, which is the lookbook operation: artworks and colors are keyed maps, so the same design routes to differently named smart objects across templates. The returned URLs are written onto draft products in Shopify and the matching listings in Etsy, with the export mode set to view so an embedded link displays the image rather than triggering a download. The team reviews products, not files.
Listing images for garments the workspace has never carried

A supplier adds a new blank and there is no PSD for it. Search POD Products finds the real garment in the print-on-demand catalog, Get POD Product returns its decoration areas, colors, and supported sizes, and those exact strings are passed into Create AI Mockup And Wait, which grounds the generated scene on that specific garment rather than an invented one. The values are matched case-sensitively, so the agent copies them from the response instead of retyping them. The new template joins a collection, and the next drop renders onto it with no further work. For a paid social slot, Create Motion Video And Wait animates the chosen still, after List Motion Models confirms which durations and aspect ratios that model actually allows.
The batch that succeeded and did not

Render Mockups In Batch answers as a success even when individual entries failed, and the failures are reported in failed_renders rather than as an error. A flow that trusts the status code publishes a product with a missing image and finds out from a customer. So the agent compares the count it requested against the count that came back, requeues the misses, and only reports a run as complete when the two agree. The same discipline applies to the AI surfaces, where a failed generation is returned in the body rather than raised, which is exactly what a flow wants to branch on. A failed video costs nothing, since the reserved credits are refunded on a terminal failure.

Human-in-Loop Highlight

Export Print Files is the operation on this connector with consequences outside the software. It produces what a print supplier manufactures from, it is billed per print file rather than per call, so a template carrying four smart objects costs four credits every time it runs, and the file it hands over becomes shirts, mugs, or embroidered caps that nobody can un-print. The setup around it is also unusually easy to get wrong in a way that looks fine. A batch render reports success while individual entries failed. Get Mockups returns only the default catalog unless told otherwise, so a template can be absent rather than missing. And image_dpi has to be set to what the supplier expects rather than left to a default.

So the agent renders, validates, and assembles the export request, then stops with the numbers on the table: "The autumn drop is 14 templates across 3 colorways. Two entries in the batch came back in failed_renders and have been requeued. Export Print Files against the remaining 40 placements would produce 62 print files at 300 DPI and spend 62 credits, and this set goes to the supplier as tomorrow's production order. Export all 40, or hold the two hoodie templates until their renders come back clean?" The production lead answers. Everything before that question ran unattended, which is the point: the agent did the placement, the colorways, the retries, and the accounting, and a person made the one call that turns pixels into inventory.

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

23 actions

Rendering

Render Mockup Renders one mockup with your artwork placed into its smart objects and returns the image URL. The core operation, addressed by the template's UUID plus one entry per smart object.
Render Mockups In Batch Renders several different mockups in one request, each with its own smart objects, and returns every result with a count of how many succeeded and failed. Read the failed list rather than assuming every entry came back.
Render Collection Renders every mockup in a collection at once, applying the same artwork and colors across all of them. The lookbook operation: one design, one call, a full product set.
Export Print Files Produces the print-ready file for each smart object rather than the mockup image, which is what a print-on-demand supplier needs. Billed per print file, and the operation this page's human gate exists for.
Create Embroidery Effect Turns a flat artwork into an embroidered-looking version of itself, which is what an embroidery listing needs before the artwork is placed on a mockup.

Templates and organization

Get Mockups Lists the mockup templates in the workspace with their smart objects and UUIDs. By default only the default catalog is returned, which reads as an empty workspace rather than an active filter.
Get Mockup Reads one template in full, including every smart object with its UUID and every collection it belongs to. This is the operation that tells a flow what a render request has to contain.
Get Collections Lists the collections in the workspace. A collection groups templates so one design can render across all of them.
Create Collection Creates a collection to group templates in. Templates are added to it in the web application or when a PSD is uploaded.
Get Catalogs Lists the catalogs in the workspace. A catalog is the outermost grouping, above collections, and is how one workspace serves several seller stores.
Create Catalog Creates a catalog and returns the UUID to pass when creating collections for that store. The vendor states this is not idempotent, so a retry produces a second catalog with the same name, and the connector publishes no operation that removes one.

Photoshop files

Upload PSD Uploads a PSD from a URL and optionally turns it into a mockup template in the same call. Without Create Template the file is stored but nothing is renderable from it.
Delete PSD Deletes a stored PSD file. Its Delete Related Mockups option also removes every template built from that file, which breaks any flow rendering them. Left off, only the source file goes.

AI mockup generation

Create AI Mockup Creates a mockup template from a generated scene or from an image you supply, so a product can be mocked up without a PSD. A prompt returns a task to poll; an image URL completes inside the call.
Get AI Mockup Status Polls a generation task. The completion signal is the body's state field rather than the status code.
Create AI Mockup And Wait Starts a generation and polls until it reaches a terminal state, returning the finished mockup. Use it when the next step needs to render on the new template.
Search POD Products Searches the print-on-demand product catalog that grounds generation. Anchoring on a real garment is what makes the output usable as a listing image.
Get POD Product Reads a print-on-demand product in full: its decoration areas, colors, and supported sizes. These are the exact case-sensitive strings a generation request has to use, so copy them rather than typing them.
List AI Styles Lists the visual styles a generated mockup can use. Pass the model so the list only offers styles that model actually supports.

Motion video

Create Motion Video Animates a still image into a short video and returns a request id to poll. An empty prompt lets the API write one from the image.
Get Motion Video Status Polls a video request. Its completion vocabulary differs from the mockup surface, which is how a flow ends up polling forever if the two are mixed up.
Create Motion Video And Wait Queues a video and polls to a terminal state, returning the finished URL. It reports a failed generation in the body rather than raising, which is what a flow should branch on.
List Motion Models Lists the video models with the durations each allows, the credit cost of every duration with and without audio, the estimated generation time, and the supported aspect ratios. Worth reading first, since a duration a model does not allow is rejected.

Frequently Asked Questions

What can FlowRunner do with Dynamic Mockups?

FlowRunner agents can run Render Mockup, Render Mockups In Batch, and Render Collection in Dynamic Mockups, plus 20 more actions.

Does connecting Dynamic Mockups to FlowRunner require OAuth?

No. Dynamic Mockups connects to FlowRunner with an API key, no OAuth flow required.

Can Dynamic Mockups trigger a FlowRunner workflow automatically?

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

Start building with Dynamic Mockups

Free plan, no card required. Connect in minutes.

Start Building Free Talk to us