You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This isn't an issue (or even really a priority at the moment), I'm just curious and thinking ahead.
With UXP development really starting to kick into gear and beginning to be accessible, and with the eventually depreciation of CEP; do you have any plans in the future to support UXP?
This repo has become so invaluable to me and my workflow and it would be a shame to be locked to a version of premiere in the future due to CEP depreciation. (plus with the benefits of UXP it'd be unfortunate to not be able to take advantage of them)
Thank you for your time, again, no pressure in any form, just curious!
The good: UXP is a more modern framework that gets rid of the client/server split and just offers a simple way to define scripts, so it should be comparably easy to work with (or at least way easier than CEP, which is like legacy by design).
The bad: Currently, UXP for Premiere is limited. Looking at the available documented functions of the Premiere DOM, it seems that CEP is still way more powerful. AFAIK, Premiere is no priority for Adobe, so this still might come.
I would postpone a complete re-development of PremiereRemote until either
a) The UXP Premiere DOM is more comprehensive and offers an actual improvement over CEP
b) CEP is not supported anymore and kicked out of Premiere, which may happen sometimes between 2026-2028, depending on how to interpret this.
Unfortunately, when I started development in 2018, I also had way more time then today (I was a student back then, now a full-time software engineer). So although I'm willing to update (or redevelop) this tooling, I might not be able to. Not the promising answer you may have hoped for, but at least an honest one ;)
So, I first did a manual feasibility study and then an automated one using Claude AI. I was worried that the UXP DOM ist still very limited, but fortunately, this seems to be wrong. This changes my perspective, and I might rewrite PremiereRemote some time later this year.
I attached the results.
Migration Feasibility Overview
Total Methods Analyzed: 75 ✓ Directly Available in UXP: 58 methods (77%) + Alternative Solution Available: 8 methods (11%) ✗ No UXP Alternative: 9 methods (12%)
Total Achievable Functionality: 88% (66 out of 75 methods)
Symbol Legend
✓ Direct equivalent available in UXP API
+ Alternative solution available via UXP documentation/workarounds
The loss of qe is definitely a tough pill to swallow, it's especially frustrating that anytime I see anyone mention a qe function and how losing it with UXP will impact them the default response from adobe is along the lines of; "well, it wasn't supported, don't understand why you'd need it, oh well". I wish they were a bit more open to hearing usecases for some of the functionality it offers/offered. ie
I believe ivan has mentioned having no access to track locking as well - shame it received no response.
I am glad to hear you're interested in providing UXP support, that's great news. I do hope adobe continues to expand UXP into the future and don't just stop short once they've achieved "backwards compatibility"
Just a heads up, I'm currently rewriting PremiereRemote from scratch to support the new UXP framework (and get rid of many many many problems the original implementation had). The UXP branch contains a funtional but very, very early version - but the core functionality of making arbitrary UXP functions executable via HTTP calls is already there :)
I finally had some free time recently to sit down and attempt to get the ball rolling with my own setup. It took me the better part of half a day just to figure out how to get the uxp plugin to actually appear in premiere (and then almost as long to figure out how to update the plugin once I've added functions or made changes lol) but once that was sorted things started working a little smoother 😅
Currently in the process of porting as many of my own functions over to uxp just to get a feel for things and I'm loving things so far!
That's great to hear! As soon as the initial development is done, I will of course invest in better documentation (well, more than 0 documentation :D). I see a lot of potential in providing a unified yet customizable interface usable both for manual and MCP-based workflows. Adobe did also some baby steps here, but far more limited than what direct UXP manipulation can reach.
Activity
sebinside commentedon Jan 14, 2026
The good: UXP is a more modern framework that gets rid of the client/server split and just offers a simple way to define scripts, so it should be comparably easy to work with (or at least way easier than CEP, which is like legacy by design).
The bad: Currently, UXP for Premiere is limited. Looking at the available documented functions of the Premiere DOM, it seems that CEP is still way more powerful. AFAIK, Premiere is no priority for Adobe, so this still might come.
I would postpone a complete re-development of PremiereRemote until either
a) The UXP Premiere DOM is more comprehensive and offers an actual improvement over CEP
b) CEP is not supported anymore and kicked out of Premiere, which may happen sometimes between 2026-2028, depending on how to interpret this.
Unfortunately, when I started development in 2018, I also had way more time then today (I was a student back then, now a full-time software engineer). So although I'm willing to update (or redevelop) this tooling, I might not be able to. Not the promising answer you may have hoped for, but at least an honest one ;)
sebinside commentedon Jan 19, 2026
So, I first did a manual feasibility study and then an automated one using Claude AI. I was worried that the UXP DOM ist still very limited, but fortunately, this seems to be wrong. This changes my perspective, and I might rewrite PremiereRemote some time later this year.
I attached the results.
Migration Feasibility Overview
Total Methods Analyzed: 75
✓ Directly Available in UXP: 58 methods (77%)
+ Alternative Solution Available: 8 methods (11%)
✗ No UXP Alternative: 9 methods (12%)
Total Achievable Functionality: 88% (66 out of 75 methods)
Symbol Legend
Premiere DOM Method Usage Table
Migration Impact Assessment
High Severity - Complete Feature Loss (9 methods)
qe.*namespace unavailablesetLockedForAllTracks,unlockTracks,applyDropShadowPreset,applyBlurPreset,applyWarpStabilizertargetAllTracks,untargetAllTracks,targetDefaultTracks,targetTracksvideoFrameRate.ticksequivalentaddCustomMarker,selectCurrentMarkerMedium Severity - Workaround Required (8 methods)
loadMarkersFromCSVFile,saveCustomMarkerToTextFilelocalFileSystemAPI with proper manifest permissionsalert()requires manifest configurationaddCustomMarkerenableAlertsfeature flag in manifestzoomCurrentClipToFitProject.getActiveProject()to verify project availabilityLow Severity - Direct Migration Available (58 methods)
createXXXAction()patternComponentChainandComponentParamTomshiii commentedon Jan 20, 2026
The loss of qe is definitely a tough pill to swallow, it's especially frustrating that anytime I see anyone mention a qe function and how losing it with UXP will impact them the default response from adobe is along the lines of; "well, it wasn't supported, don't understand why you'd need it, oh well". I wish they were a bit more open to hearing usecases for some of the functionality it offers/offered. ie
I believe ivan has mentioned having no access to track locking as well - shame it received no response.
I am glad to hear you're interested in providing UXP support, that's great news. I do hope adobe continues to expand UXP into the future and don't just stop short once they've achieved "backwards compatibility"
sebinside commentedon May 14, 2026
Just a heads up, I'm currently rewriting PremiereRemote from scratch to support the new UXP framework (and get rid of many many many problems the original implementation had). The UXP branch contains a funtional but very, very early version - but the core functionality of making arbitrary UXP functions executable via HTTP calls is already there :)
Tomshiii commentedon May 14, 2026
Hopefully I get some time in an upcoming weekend to have a play with it!!
Tomshiii commentedon Jun 21, 2026
I finally had some free time recently to sit down and attempt to get the ball rolling with my own setup. It took me the better part of half a day just to figure out how to get the uxp plugin to actually appear in premiere (and then almost as long to figure out how to update the plugin once I've added functions or made changes lol) but once that was sorted things started working a little smoother 😅
Currently in the process of porting as many of my own functions over to uxp just to get a feel for things and I'm loving things so far!
sebinside commentedon Jun 21, 2026
That's great to hear! As soon as the initial development is done, I will of course invest in better documentation (well, more than 0 documentation :D). I see a lot of potential in providing a unified yet customizable interface usable both for manual and MCP-based workflows. Adobe did also some baby steps here, but far more limited than what direct UXP manipulation can reach.