It would be nice to have a button, in ServiceStudio, to remove all unused references in an eSpace, or a list of unused references which we could then remove.

Hi, I found that 100 espace limit in a page was too few.I usually maintained Platform Server once in a few months.And it really a chore when deleting old versions of espace.Maybe add a configurable paramater instead of static 100 entries.

Sometime you work on a certain part and are not able to finish certain parts, or want to skip certain parts, like complex queries.Then it would be great to disable or, comment out certain tools in the action flow. so you don't have to worry about them now.Perhaps they are not complete yet, or unusable.With an if-block, it would always follow the true-line, it would skip the for-each etc. etc.

Looking at the authorizations offered while creating the connection to MCP (Model Context Protocol), all the options related to (OAuth) ask for a client secret as a mandatory field. While searching on MCP Documentation, the client side must implement a PKCE (Proof Key of Code Exchange) flow, with this flow its not mandatory to have a client secret to get Authorization to interact with the MCP Server. Making a lot of MCP Servers impossible to connect to OutSystems. Where is the link for the documentation where they explain why PKCE flow should be mandatory: https://modelcontextprotocol.io/specification/2025-11-25/basic/authorization#authorization-code-protection My suggestion here is to make client secret not mandatory.

Currently, if we define a Sorting by an attribute, that sorting becomes unresponsive after introducing a Group By in the aggregate, as the sorted attribute is no longer part of the aggregate output . This does not raise any errors or warnings in the TrueChange Tab or any signaling in the workspace. I find that this is an easily overlooked situation that can lead to unexpected behavior or inconsistent results. I suggest the implementation of a typical Outsystems background validation that triggers a warning (written on the TrueChange Tab and a visual cue on the aggregate) whenever the aggregate's sorting references attributes that are not part of the aggregate output after a Group By, ensuring consistency and preventing silent issues.

Allow server actions and integrations (e.g. AI use cases) to stream text to the UI instead of waiting for the full response. This is standard on any AI powered app and since OutSystems is pushing AI features, I think this is a must have. We don't want our users to have to wait for the full LLM response before having some feedback. A way to implement this is to allow Server Sent Events which would even facilitate other use cases, especially in ODC with Event Driven Architectures.

Hi everyone, I’ve noticed something with the "Duplicated Code" pattern in AI Mentor that has been causing a lot of false positives in real projects. Right now, the detection seems to focus mostly on the Action flow structure (basically the “shape” of the logic). The problem is that many Actions in OutSystems naturally share the same simple pattern, like: Start → SQL → Assign → End So even when two Actions have completely different SQL queries and return different Output Structures, AI Mentor still flags them as duplicated just because the flow looks the same. In practice, these aren’t duplicates at all. They just follow the same architectural pattern. Trying to merge them usually doesn’t make sense, can hurt readability, and sometimes even impacts performance. Suggestion It would help a lot if the duplication check also considered the SQL widget details and the Action contract, not only the flow. For example, avoid flagging when: the SQL statement text is clearly different the Output Structure is different the returned data shape is different Maybe duplication should only be reported when the flow and the SQL and the outputs are very similar. This would make the detection more semantic instead of purely structural. Benefits Reducing these obvious false positives would: • Reduces false positives • Less time wasted marking valid code as “Won’t Fix” or "False Positive" • Prevents unnecessary refactoring that may hurt performance • Increases developer trust and adoption of the tool I think this small change could make the Duplicated Code pattern much more accurate and much more valuable in day-to-day development. Thanks!

It is great that the OutSystems demo Homebanking apps are multilingual and also available in German. Nevertheless it is obvious the translation was not done by a German native speaker. In this way it is - to be honest - a little bit funny, instead of being a serious business app. E. g. the credit card VISA was translated as Visum in German language, which only means a visa for a country, not the credit card. And a lot of other stuff. English name -> curent translation - suggested translation Visa -> Visum -> Visa (I think it the same for the Portuguese word here: Visto) Home -> Heim -> Startseite Balance -> Gleichgewicht -> Kontostand Amount -> Menge -> Betrag Continue -> Weitermachen -> Weiter Loans -> Loans -> Kredite View -> Sicht -> Überblick Share Receipt -> Aktienquittung -> Quittung It would be great if these Homebanking apps (maybe others also) could be checked by a native German speaking OutSystems employee of the DACH-region.

Currently, OutSystems only allows limiting the total number of simultaneous process executions across the environment. There is no way to restrict concurrency per process type, which can cause problems when a single process spawns many instances. For example, starting 1000 BPT processes of the same type can consume all available execution slots, effectively blocking other processes from running. A per-process concurrency limit would allow developers to define a maximum number of simultaneous executions for each process (e.g., 4 instances at a time). Additional instances would automatically be queued until a slot becomes available. This feature would provide: Better control over resource usage for high-volume processes. Reduced risk of process starvation, where other processes cannot execute because one process is using all slots. Simpler queue management without custom throttling logic. Introducing per-process execution limits would make process orchestration more predictable and efficient, allowing developers to safely run multiple processes in parallel without one process dominating the system.

Currently, when using an Input widget with Type = Number, it is not possible to set a Max Length for the field. This forces developers to switch the input type to Text and implement additional JavaScript, masks, or custom validations just to control the number of digits — which increases complexity, reduces maintainability, and removes the native numeric keyboard on mobile. Adding a Max Length property (similar to text inputs) would allow developers to: Prevent users from entering more digits than required Avoid using JavaScript workarounds Keep the input type as Number Improve UI validation consistency Increase development speed This enhancement would significantly reduce unnecessary logic and provide more control over numeric inputs without losing native browser behavior