Integration · Shopify

Generate on-model photos for your Shopify catalog

Start with your assistant and the Product page set template. Use your garment photos to create studio views, then add them to the right product.

Model wearing a suede zip jacket, full-body front on the Pure White studio, generated with nano-banana-pro through the Product page set template
Front hero from the Product page set template: suede jacket on the Pure White studio, generated with nano-banana-pro through the Uwear MCP.

Use cases

What can you build?

  • Automate on-model product photos

    Generate front, upper-body and back views plus a 360 video per garment. A video model draws the sides that the source photo does not show, so check the clip against the product. Map the result URLs to Shopify product IDs and append them as product media.[25][30][58]

  • Build a virtual try-on storefront experience

    Prototype garment selection with Uwear MCP. Build the shopper interface as a Shopify theme app extension, with an app backend that runs Uwear API jobs and keeps personal previews out of public product media.[45][47]

Model wearing the same suede zip jacket, full-body back camera edit on the Pure White studio, generated with nano-banana-pro through the Product page set template
Back view from the same Product page set run: a camera edit of the front hero, so the model, light, and studio carry over. Generated with nano-banana-pro through the Uwear MCP.

Assistant workflow

How do you run this from your assistant?

Uwear ran this workflow for one product on a Shopify development store on September 18, 2026, with the Shopify connector for Claude and Uwear MCP in the Code tab. Read the field notes for the results and limits. This is not a preinstalled connection. The chat tab and other assistants were not tested.[11][12][13][57][58]

  • Shopify plugin for ChatGPT / Shopify connector for Claude / Perplexity

    Use Shopify’s merchant connections in ChatGPT, Claude or Perplexity. Claude and ChatGPT document product reads and image updates; Uwear tested the Claude connector in the Code tab. Check each host’s tools and permissions.[11][12][13][57][58]

    Public installation docs; store login and access approval required.[11][12][13][57][58]

  • Dev MCP server / Shopify AI Toolkit

    Supply developer docs, schemas and validation. Install the Toolkit in Codex with codex plugin add shopify@openai-curated. Store writes use the separately authenticated Shopify CLI store execute command with --allow-mutations.[8][9]

    Local Dev MCP; authenticated CLI execution for store operations.[8][9]

  • Storefront MCP / Storefront Catalog MCP

    Support shopping tasks and product discovery. The documented tools do not write product media; use a merchant connector or Admin GraphQL for that step.[7][17]

    Public implementation docs; no explicit GA or beta label.[7][17]

  1. Connect Shopify and Uwear

    Connect Shopify and Uwear MCP in a supported assistant. Shopify lists ChatGPT, Claude and Perplexity; check both connections in your chosen host. Uwear tested the Claude Code tab. The Shopify AI Toolkit and authenticated CLI offer another coding-assistant route.[8][9][11][12][13][57][58]

  2. Select products and keep their IDs

    Read source photos, product IDs and variants with search_products and get-product. Keep each Shopify ID linked to its Uwear garment and result.[11][12]

  3. Approve the Product page set brief

    Ask Uwear MCP to load Product page set for the selected garment photos. Review the art direction and credit estimate before generation. Automatic QA compares results with source photos and retries flagged results when switched on.

  4. Append the generated media

    Retrieve the Uwear image URLs. Simple connector actions keep the source file name; use fileCreate through graphql_mutation with filename and alt for readable names. Wait for READY, then attach with fileUpdate. A hosted video needs stagedUploadsCreate and a multipart byte upload, which a coding assistant can send from a shell; its external URL alone will not work.[12][25][26][29][30][58]

  5. Check the product gallery

    Read each product back by ID after a write. Check images and alt text with get-product; query product { media } for video. Wait for processing and read back the order after productReorderMedia. Set the intended variant preview and check the storefront before publishing.[25][5][58]

Run it from your assistant

Run this photoshoot from your AI assistant with a Uwear template.

Connect the Uwear MCP server to ChatGPT, Claude, or Codex and ask for the Product page set template. It is a batch workflow that runs once per garment: front, upper-body, and back product-page shots on a white studio background, plus a 360-degree video, from one garment photo. Any Uwear account can load it by name, and the assistant shows a costed brief in credits before generation starts.

Copy this prompt into your assistant

Connect the Shopify plugin for ChatGPT or the Shopify connector for Claude and the Uwear MCP server in this session. Read my tagged Shopify products and keep their product and variant IDs. Load the "Product page set" template for their garment photos on its white studio setup, turn the AI reviewer on, and show me the costed brief in credits before generating. After I approve, use fileCreate through graphql_mutation for readable filenames and alt text, wait for READY, and attach with fileUpdate. Use stagedUploadsCreate and a multipart byte upload for video if this session can send the bytes. Check the clip against the garment photos. Read each product back by ID after a write, query product { media } for video, and verify the gallery before reporting.

What one run does, per garment

  1. Generate the front hero

    One full-body front photo of the garment on the template model, in the Basic white photoshoot art direction on the Pure White studio location.

  2. Edit to the upper-body crop

    A camera edit that reframes the front result to an upper-body front frame instead of generating a new image, so the crop comes from the same session as the hero.

  3. Edit to the back view

    A second camera edit of the front result to a full-body back frame, so the product page gets the back of the garment from the same session.

  4. Render the 360 turn

    A 5-second Kling 3 Pro video from the front photo: the model turns a slow, full 360 degrees in place, garment readable throughout.

Outputs per garment

  • Full-body front photo, 2K, 2:3
  • Upper-body front photo, 2K, 2:3
  • Full-body back photo, 2K, 2:3
  • 5-second 360-degree video

The setup the template pins

Template kind
Batch workflow: runs once per garment row
System key
demo_full_pdp_test_white_studio
Art direction
Basic white photoshoot (system art direction)
Location
Pure White (system studio location)
Model
The Uwear system model by default; the assistant can swap in one of your saved models when it proposes the brief
Image model
nano-banana-pro, 2K, 2:3
Video model
Kling 3 Pro, 5 seconds
QA
Optional. With the AI reviewer on, each result is checked against the source product photo and flagged results are retried
Approval
The assistant shows a costed brief in credits before anything generates

Pure automation

How does API delivery work?

Send images by public URL. Simple connector actions keep the source file name; for readable names, use fileCreate through graphql_mutation with filename and alt. Wait for fileStatus READY, then attach with fileUpdate. A hosted video cannot be added by its external URL: use stagedUploadsCreate and a multipart byte upload, which a coding assistant can send from a shell. Assign a variant preview after attaching the image.[25][26][29][30][58]

  1. Start from a product event

    A service receives Shopify webhooks, imports the garment into Uwear and keeps the product mapping. Deduplicate events and ignore changes caused by your own media delivery.[56]

  2. Run generation without an assistant

    Submit Uwear REST API jobs and receive completion webhooks. Or start a published Production Automation after a garment is processed, a tag is added or an outfit is created. These are Uwear workspace triggers.

  3. Deliver through Admin GraphQL

    Your service imports the result, waits for readiness and adds the Shopify product reference. Production Automation still needs this API delivery step; no native Shopify publishing node is assumed.[25][30]

Platform requirements

What image specifications apply?

Shopify: What image specifications apply?
SpecificationRequirement or guidance
Product image sizeUp to 5000 × 5000 px or 25 megapixels; keep the file smaller than 20 MB. These are product-media limits, not theme-upload limits.[35]
Recommended composition2048 × 2048 px for square product images; use a consistent aspect ratio for featured images. Square is a recommendation, not a requirement.[35]
Files upload formatsJPEG, PNG, WEBP, HEIC and GIF. Validate the format through your selected upload path.[36]
Gallery and variant viewsUp to 250 media items per product, including images, video and 3D media.[4]
Variant previewOne assigned image per variant. Keep extra views in the product gallery.[5]
Hosted videoMOV, MP4 or WEBM; maximum 1 GB and 10 minutes. The Product page set clip is 5 seconds.[36]
Alt text and filenameUse filename and alt when creating the file. Describe the garment and view; filenames can receive a unique suffix.[29]

Existing options

Which AI image tools are already available?

Documented features from the linked sources. These are not independent quality tests.

  • Shopify Magic / Sidekick

    Built-in background and prompt-based image edits suit simple changes in the file editor. Check feature access in your account.[38]

  • Botika Connector

    Imports Shopify products into a Botika account and sends Botika images back. A fit for teams already using Botika that want a packaged product transfer.[40]

  • Photoroom AI Listing Assistant

    Lists virtual models, backgrounds, bulk editing and Shopify publishing. A fit for product photo cleanup and listing maintenance.[41]

Frequently asked questions

Yes. On September 18, 2026, Uwear executed this route for one product on a Shopify development store with the Shopify connector for Claude and Uwear MCP in the Code tab. Shopify imported generated images by public URL. Simple connector actions kept the source file name; fileCreate through graphql_mutation supplied readable names and alt text, followed by READY checks and fileUpdate to attach the images. Read each product back by ID after a write. Other assistants and the chat tab were not tested.[11][29][30][58]

Yes, with a staged upload. In Uwear’s September 18, 2026 test, fileCreate rejected the external Uwear .mp4 URL with Invalid video url. A coding assistant used stagedUploadsCreate, sent the bytes by multipart POST with curl from a shell, then called fileCreate with resourceUrl, waited for READY and attached the video with fileUpdate. Read product { media } to verify it; get-product lists images only. A chat assistant with a networked code sandbox was not tested.[26][29][30][58]

Shopify recommends 2048 × 2048 px for square photos and consistent featured-image aspect ratios. Square is not mandatory. The Product page set template uses 2:3 photos; check the crop in your theme before publishing.[35]

Prototype the garment mapping and generation with Uwear MCP, then build a theme app extension connected to your backend and the Uwear API. Your backend handles shopper consent, photos, job status and usage limits. Keep personal try-on results separate from public product media.[45][47]

This proposed integration uses an API service to map Uwear results to Shopify products and attach ready media through Admin GraphQL. Uwear triggers start generation inside the workspace; the service owns the Shopify delivery step.[25][30]

Uwear uses credits at $0.10 per credit, with model rates on the Models page. Your assistant shows a costed brief before generation. QA is automatic and toggleable: when on, an AI reviewer compares results with source photos and retries flagged results.

Yes. Generate in Uwear Studio, download the results, then upload or select them in the Shopify product’s Media section. Choose the intended featured image and variant preview there.[4][5]

Research

Sources and access dates

  1. [4]
  2. [5]
  3. [7]
  4. [8]
  5. [9]
  6. [11]
  7. [12]
  8. [13]
  9. [17]
  10. [25]
  11. [26]
  12. [29]
  13. [30]
  14. [35]
  15. [36]
  16. [38]
  17. [40]
  18. [41]
  19. [45]
  20. [47]
  21. [56]
  22. [57]
  23. [58]

Next step

Connect your catalog to Uwear.

Start with an assistant-led test, then build the API delivery your catalog needs.