The problem
Testing Luna meant recording firmware builds, then opening a separate compression app before I could ask AI to review each video. PromptCam grew out of wanting to remove that repeated step.
The intended workflow
Capture or import a photo or video, mark the detail that matters, reduce the file size, then copy, share or download it. I want to keep those steps together instead of moving through several apps.
For a Luna test video, the important detail might be text on the screen, a change in state, or a problem with a response. A smaller file is only useful if that detail survives. The question is how small the file can be while still showing what I need help with.
What needs to work
The whole flow matters: a ready camera preview, an obvious recording state, a Stop control within reach, and a way to get a playable file into the next tool. Saving a recording is only part of the job if I still have to troubleshoot sharing or compress it somewhere else.
Where it stands
PromptCam is still in development. The video-repair candidate has passed automated checks, but it still needs a real-phone test. Those checks do not establish that the capture-to-share experience works on a phone.
The next milestone is an end-to-end test: record something I need help with, keep the important detail, and get a playable file into the AI tool I intended to use. Full-length imported-video compression and time-range video annotation remain planned work.