This is kinda useless, but also, what do you MEAN that we couldn’t do this before? What did plugin windows do so differently in the studio UI code that they couldn’t get screenshotted? And why did you change it now?
From what it seems, the main objective of this is to enable MCP support for screenshotting specific windows so eventually, AI can develop plugins more efficiently with a direct view of what its working on (the plugin window).
first of all, this is off topic, please don’t post things off topic to announcements and also, this should already be possible, if i’m understanding your request. I have a project that entirely relies on editing captures and it works in production.
There still isn’t EditableImage support for CaptureService
Please do give me whatever magical code you have that allows editing of captured screenshots.
Though I noticed it doesn’t work when focused on the script editor, which is most of the time for me. X)
Is there a chance that it could get resolved? Thank you!!
Can this support capturing the full contents of a scrollable window?
This could be useful for reporting Assistant bugs, but only if it can actually capture the entirety of the logs at once, yet even that would result in huge images, so I’m not sold this is the right avenue for log-reports.
Right now, allowing the agent to take control of the computer is the only reliable way to diagnose visual bugs. Allowing it to debug without interrupting everything else is the missing link!