We are excited to share the release notes for Sprint #96 of Icecat Studio. In this continuous deployment cycle, our team focused on three critical areas. We reinforced platform-wide security protocols and improved analytics dashboard performance. We also optimized the architectural backbone for upcoming AI-driven tools.
Let’s dive into the details of what was delivered in Sprint #96.
Security remains our top operational priority. In this sprint, we have officially rolled out a mandatory Two-Factor Authentication (2FA) enforcement flow across the platform. High-privilege profiles, such as editors, superusers, and supereditors, now face automatic restrictions upon login. The system guides them to a dedicated 2FA setup screen to protect administrative layers from unauthorized access.
To maintain a frictionless user experience for commercial brand users, we implemented an automated, once-daily skippable flow. Brand users can safely bypass the activation prompt with a dedicated “Skip” button, proceeding directly to their active content workflows without unexpected task disruption.
If you’re an Icecat Studio user, please don’t forget to turn your 2FA on!
Navigating dense visual components in the editor can slow content delivery. To eliminate this friction, we deployed a highly requested drag-and-drop structural update at the module group level in our administration settings.
System administrators can now control the order of components within categories. This allows teams to pin high-demand modules at the top of the interface. Additionally, components can now dynamically map to multiple parent groups simultaneously. A clean multi-select field supports this workflow. This update removes structural duplicate items from our library asset management pipelines.
Moving another step closer to a hands-free, autonomous product catalog, we have continued releasing background engineering updates to our automated ingestion modules. This sprint lands baseline FTP connection pipes and optimized feed preset mapping configurations. The engine silently parses raw supplier product sheets in the background, minimizing manual asset updates and ensuring content stays up to date.
Our upcoming AI Agent has advanced from initial blueprints into deep engineering execution. During Sprint #96, our team conducted targeted technical investigations, structural planning, and configuration of the core backend architecture.
We successfully established a streaming-capable WebSocket gateway module to handle real-time AI turns safely and laid the foundation for future releases, bringing us closer to usable prototypes.
We know that real-time data is crucial for assessing the returns of marketing campaigns. This sprint, our team ran a rigorous data-layer optimization sprint focused on the heavy analytical and event telemetry aggregation layers within our ClickHouse infrastructure.
By reindexing historical calculation passes, we significantly reduced loading times on the brand dashboard. We have also investigated and begun planning future updates and a global refactoring of the dashboard, which will positively impact performance and the usefulness of insights.
To give brands full ownership of product launch dates and promotional timelines, we have shipped database migrations that support independent publish_at and unpublish_at nullable timestamps, backed by an hourly background cron daemon. Initial development is limited to the Icecat Network syndication channel while skipping the external Amazon API. The next step will be UI implementation of the scheduled publication tool.
We have decoupled hardcoded content precedence rules by launching an autonomous Variant Priority Order schema on the home Brand Settings interface. It’s now possible to set up a global drag-and-drop fallback list across our primary story variants. The distribution engine follows a cascade logic resolution: it respects specific channel “Publish rules” first, but instantly uses the brand’s fallback priority order to serve content if no specific category or locale condition is found.
We have laid a foundation and engineered a dedicated database collection, along with an extended execution engine, to support secure, mass status updates across dynamically filtered asset sets. The new collection systematically records metadata, including action types, search filters, progress metrics, and execution error logs, in real time. This architecture ensures absolute data integrity and provides comprehensive status reporting for large-scale catalog syncs.
Balancing creative design freedom with strict compliance guidelines is always a challenge. To solve this for Amazon standard layouts, we updated the Story Builder behavior to open inline content editing access while strictly guarding global visual consistency. Users can freely adjust texts, translate paragraphs, and modify media data nodes. However, style-related configuration attributes for critical structural elements—such as Premium Comparison Tables and Navigation Carousels—are strictly locked at the entity property level, ensuring approved layouts remain safe from breaking style alterations.
Our completely overhauled web rendering standard is rapidly nearing production readiness. This sprint saw the framework successfully deployed to active Dev and QA environments for continuous cross-browser testing. Built entirely on native Web Components and Shadow DOM root encapsulation, the new engine isolates Icecat layouts entirely from external noise. It guarantees that our rich media layouts never break a merchant’s global CSS layout rules and drastically reduces the bundle footprint to optimize mobile Core Web Vitals.
To eradicate unexpected storage performance bottlenecks under heavy data influxes, we have unified our local infrastructure and external media repositories into cohesive storage pools. We kicked off active beta testing to migrate historical local Minio asset directories to cloud-managed, external S3 storage instances provided directly by our global CDN partner, Bunny. This integration prevents system downtime caused by disk space limitations and ensures rapid asset loading for international end consumers.
Our visual AI prototyping space has received vital upgrades to user interaction based on editor feedback. Images imported directly into an open module canvas are now recognized as layers rather than background images, allowing for full transformation. If needed, the layer can still be set as a background.
Furthermore, clicking the “Set as background” action now accurately anchors the image’s position while fully preserving your custom scaling coordinates and cropping aspect ratio, without unexpected viewport resets. Previously, the image fit the canvas, which was not always the expected result.
Crop, Fill, and Fir options are still available through the “Size” property on the “Properties” sidebar.
If you are interested, click here and book a demo with Icecat Studio.
Read further: Icecat, e-commerce, ecommerce, Icecat, Icecat studio, product content, release notes
