Icecat Studio – Sprint 92 Release Notes

Sprint 92 marks a major functional leap for Icecat Studio with the debut of our AI Image Assistant MVP, empowering brands to generate professional-grade visual content through conversational AI. Alongside this innovation, we delivered significant architectural refactors to Component Groups and the Product Story Framework v2, while simultaneously enhancing the Submission Process UX and scaling our Live API performance through CDN integration.

AI Image Assistant MVP (Sandbox)

The Sandbox – our embedded image editor, now has an AI assistant that can generate background images. At the moment, it’s available to all users, so feel free to explore and experiment.

This feature is currently available to all users and is especially valuable for smaller and mid-sized brands that want to create more polished visual content without requiring extensive design resources.

Key highlights:

• The AI generates backgrounds only – it does not modify product images or text layers. Your product remains untouched (no random buttons or extra artifacts).
• It understands all layers on the canvas, so you can take that into account when writing prompts.
• It supports conversational context – you can iteratively refine results:

“generate a kitchen for my coffee machine → add a cup of coffee → change the table color → replace the cup with a pint of beer.”

• “Apply background” button lets you instantly preview the generated result on your banner/image.
• We also expose the model’s reasoning, so you can see how the AI approaches image generation and refine your next prompt accordingly.

---

Submission Process Progress Bar Improvements

We’ve introduced a UX update to the Submission Process Progress Bar to make the publishing flow more transparent, interactive, and easier to navigate.

What’s new:

• Each submission status is now clickable
• Users can jump between steps when the process is already in progress (for example, go directly to Publish)
• Added animations and hover indications for clearer interaction cues
  Introduced a new Published status, so users can clearly see when a product is live
• Added process information, definitions, and tooltips to improve transparency of each step
• The Publish status now gives a broader indication of which channels the story is published on:

These improvements were inspired by feedback gathered from brand users and are part of our ongoing effort to simplify and enhance the publishing experience.

---

Component Groups

Our team has refactored Component Groups from a hardcoded constant into a fully dynamic, database-driven structure.

Previously, component groups were stored as static constants in code, which meant any changes – such as adding, editing, reordering, or deactivating groups – required development effort and code updates. As part of this improvement, component groups were extracted into a dedicated MongoDB collection with full CRUD support via backend and API layers.

What changed:

• Component groups are now stored in a dedicated database collection
• Groups can be managed dynamically (create, edit, reorder, activate/deactivate, delete)
• Components now reference groups via database IDs instead of static numeric values
• Existing component data was migrated from hardcoded group IDs to the new structure
• API responses remain backward-compatible by still returning the full group object in component responses

This makes the system fully manageable from the admin side, easier to maintain, and ready for future scaling without requiring code-level changes.

---

Import Interface – Reporting UI

Continued development of the Import Interface Reporting UI for the External Gateway, building on the foundation introduced in Sprint 91.

This sprint focused on further refining the reporting experience for brand users, improving the visibility of import outcomes, and strengthening the UI layer around import monitoring and troubleshooting.

As introduced previously, the feature includes:

• a dedicated Import section for brand users
• visual indicators for import status and processing outcomes
• downloadable reports at both batch and asset levels
• integration into the broader Report Log experience

This ongoing work supports more transparent external feed onboarding and helps users identify and resolve content processing issues more efficiently.

---

Story Builder v2 – Left Sidebar: Modules Library & Module Tree

Work on the new Left Sidebar for Story Builder v2 has progressed, introducing a host-driven interface for both adding modules and managing story structure.

We have established the new sidebar foundation and introduced the Modules Library tab as the main entry point for browsing available modules and adding them to a story. Also added the Tree tab, which provides a clear structural overview of the story in a host-side interface. Users can now:

• view modules in story order
• reorder modules via drag-and-drop
• trigger key module actions directly from the sidebar, such as:
  • edit
  • duplicate
  • delete
  • open AI-assisted content generation
  • open decorators/styles

Additional improvements include:

• centralized handling of AI-related actions across UI entry points
• better action hierarchy and accessibility
• consistent hover/active states for module items
• action restrictions

In short, these developments lay the groundwork for the new left sidebar in Story Builder v2 and cover two key workflows: adding modules and managing story composition in a more structured and scalable way.

---

Assets Group Actions

Built the foundation for Assets Group Actions – a more scalable way to perform bulk operations across selected groups of assets.

Instead of creating a one-off solution for a specific operational need, we introduced a more flexible approach that uses existing Search API filters to dynamically identify relevant assets or story variants and apply actions to them in bulk.

This creates a stronger long-term foundation for repeatable operational workflows, including future bulk actions such as publish, unpublish, and other large-scale content management tasks.

---

Product Story Framework v2

Made significant progress on Product Story Framework v2, which is now nearly complete.

The framework has been substantially updated and effectively rewritten for a new architecture, resulting in better stability, improved performance, and stronger scalability.

At the same time, it is already clear that the next step – integration with the Builder – will require additional work, particularly regarding compatibility with the builder environment and proper interaction through Shadow DOM/Shadow Root.

In short, the v2 framework is largely ready from an architectural perspective and already delivers clear technical improvements, with Builder integration being the next major milestone.

---

Set up CDN for Live API

The Live service now operates through a CDN layer. This architectural change improves content delivery speed for end users, reduces the direct load on our backend servers, and ensures greater stability for live content distribution.

---

If you are interested, click here and book a demo with Icecat Studio.