Today is only possible to search an application by name. But would be nice to have some kind of filtering too. Today I missed a way of filtering by type (web, mobile, service). Also for example hide the applications that came from forge. But for sure there is a lot more filtering that we can use. Would be nice to discuss around this topicAnd image to be clear the screen I'm talking about:

I observe that the people who are new in Outsystems they complete the course but when it comes to certification exams they found question in exam confusing and they perform poor in exams.  I think there should be at least one mock test before certification.

We've seen a lot of discussion around Static Entities lately, for example, this idea . While Static Entities provide some value for enumeration use cases, they fall short in a lot of ways: They are stored in the database. They are expensive compute wise (more than they should be) to retrieve. They cost AOs. To use them on client side (other than the ID), they need to be manually synced. Developers need to understand the aforementioned details to use them correctly. Traditional Enums in high code languages are much better. They are directly in the code, making them available at no cost to the caller. Since OutSystems is a giant compiler, Enums should be a first-class construct that: Are compiled into constants rather than stored in the database. Are accessible both server-side and client-side without requiring manual synchronization. Do not consume AOs. Can be used in Service Studio like a variable. Maintain type safety like a Static Entity ID. Static Entities are an old primitive of the platform. I don't believe changing them at this stage is a good move, so I propose creating a new primitive—Enums—that align with modern best practices while keeping the low-code philosophy.

It would be great to have a Start Index  property in Aggregates so that I can have pagination patterns even without UI. This way, I could state that I want 50 records starting from record 50 (e.g. to display them in the second page of a table).

In the new Service Studio alert that shows the forge components that have new updates, there could be an "update all" option or checkboxes and and an "update selected" option.

 Context: When a merge is performed and copy/paste is detected, the system should notify the user immediately. This notification is crucial so the developer understands that the original reference will be lost in subsequent environments — TST, HMG, PPRD, and PRD — with production being the most critical. Entity Risk: Data Loss : Copy/paste can overwrite or eliminate important references, leading to inconsistencies and missing information. Process Impacted : Suspension of Active Processes : Losing references may interrupt workflows or cause unexpected behavior in critical environments. Current Problem : At present, the merge process does not provide a clear “red alert” warning when this situation occurs. As a result, unaware/new developers may unintentionally commit this error, causing data loss and instability. Proposed Improvement: Preventive Alert System Automatically detect copy/paste during merge. Display a high-visibility warning (e.g., “⚠️ Warning: Original reference will be lost”). Developer Awareness & Accountability Require explicit confirmation before allowing the merge to proceed. Log the incident for traceability and quality assurance. Benefits Prevents data loss across environments. Improves system reliability in production. Ensures developers are fully aware of the consequences before committing changes.

Currently, the only way to share Outsystems code with LLMs or to show to a person on the forums, etc, is with a Screenshots, but a Screenshot cannot display everything at once: for example Input/Output parameters for several Actions in an Action flow in a single Screenshot. A single one side Service Studio -> Text Editor would open new doors, if you can Copy/Paste Service Studio elements in a text form to paste into LLMs or other text editors for AI code generations , context, creation of documentation, etc. It doesn't have to be perfect, it can be an imperfect, incomplete and invalid machine generated code (JSON, XML or any other text form, that is easier for the Service Studio team to implement), LLMs nowadays can quickly reformat it in a better form according to use case and extract whatever needed. It can help by virtually all of a module/application elements inside Service Studio, but among the most beneficial would be StaticEntities and Entities + attributes, Structures + attributes, Action in a flow (names + input/output parameters), etc.) in a text form. It is not the same as, but it would implicitly provide some solution for the problem from this idea: https://www.outsystems.com/ideas/11114/copy-multiple-property-names/

When developing in Service Studio on a UI and when using a popup widget, this widget can be made visible or not (Display mode: show all placeholders or hide placeholders). Nevertheless it can be a bit frustrating when working on the screen and a popup at once. This since the popup always hovers over the screen in service studio. The idea would be to create a new display mode for the popup, so that the popup can be edited, but is placed on the bottom of the screen. Just like a sidepanel can be edited. So the screen and the popup can be edited without switching the display mode of the popup. This will increase the development time on the UI.

Currently, developers receive community points for submitting Forge components from OutSystems 11, but no points are awarded when submitting from ODC. It would be great to align this with O11 and grant points for ODC submissions as well. This enhancement would motivate developers to contribute more reusable Forge assets in ODC, fostering greater community engagement and innovation.

It would be incredible if we would be able to add the cache_control header in the messages we are sending to the LLMs using the new AI Workbench. This feature is available at least on Claude with more info here: https://docs.claude.com/en/docs/build-with-claude/prompt-caching This approach, and I quote, "significantly reduces processing time and costs for repetitive tasks or prompts with consistent elements." This allows companies substancial savings on token usage! Looking at the documentation they clame that this is especially useful for: Prompts with many examples Large amounts of context or background information Repetitive tasks with consistent instructions Long multi-turn conversations where I would underline the repetitive tasks, which is one of the advantages of using agentic AI, where it can call the same tools again and again, with the same system prompt. In the documentation the example they give is to cache an entire book. In our case, these could be large documents with company details, or a quite big system prompt.