Hello Johnny,

Thank you for your review and feedback, greatly appreciated.

I’d like to share some details about our current features and what we’re working on next in this area.

At the moment, Robomotion does not include a full desktop recorder, as you mentioned. We currently provide Inspector Tools for identifying UI elements, along with Image Automation actions that allow you to capture images manually. This means you need to inspect elements and capture images one by one for each action, as shown in the video below.

Just as a note: for applications that do not expose standard Windows UI controls, every RPA tool faces the same limitation. If an application does not provide accessibility properties, Windows UI Automation cannot interact with it, and the only available method becomes Image Automation.

You can verify this using Microsoft’s own Accessibility Insights tool:
https://accessibilityinsights.io/

If Accessibility Insights cannot detect or highlight an element, then no UI Automation library will be able to automate it. In that case, each step must be captured manually using image-based actions.

Here is our current Inspector Tool, which I believe you’re already using:

https://packages.robomotion.io/releases/tools/robomotion-windows-inspector_v0.1.1.zip

and here is how it works, for readers:
https://www.loom.com/share/06a3d381721544e5b20413bfbf042285

We do plan to unify all the inspector tools we use in our automations, all for web pages, windows desktop apps, windows java apps, image automation, sap automation (which is currently actually sap's own inspector tool we recommend and use) into a single universal inspector and improve the automation and recording experience in the future. Your feedback helps us prioritize these improvements.

But right now, we have two major priorities. The first is the Agent Store, which we will open to community developers so they can contribute and monetize their agents and domain expertise. The second is AI-assisted, prompt-based Flow Generation, where you will be able to build flows through iterative conversations, similar to how developers work with tools like Claude Code. We aim to deliver this feature in a couple of months, hopefully in January.

We are also thinking and trying to find ways to integrate this AI assistance into the universal inspector so that during image, web, or desktop automation, it can guide you, provide suggestions, and help validate actions as you inspect or record. These features are part of our future roadmap, and we aim to deliver the universal inspector next year to provide a better and easier automation experience.

Best