Thanks for sharing.

Looking at your projects brought a couple things to my mind. I use the agent to make the initial cuts removing silence, etc, from the transcript; so when the project is loaded it already is "half" baked. Then I just need to adjust the cut.

Do have a similar marketing workspace for my projects, one for each, with a abstract repo that have all skills and brand assets, tone of voice etc. By starting a new project the new workspace initialize by interviewing the user to gain all information of the project/product and then start defining assets, tone of voice, workflows, etc. Been using that for several month now, it gets better as the models get better.

Wondering what I need to adjust to open source it...