phongmenly/editvideoskill

This project is called video-brand-kit, though the repository itself is named editvideoskill. It is a small companion tool built to work alongside a separate project called video-use, which handles the core work of editing talking-head videos: transcribing speech, cutting clips, color grading, and rendering the final file. This kit does not replace or fork that tool. Instead, it adds one creator's specific visual style on top of it, including a neon yellow and white color scheme, short keyword callouts shown on screen instead of full subtitles, a circular picture-in-picture layout for longer animated segments, and a branded opening hook shown at the start of a video. The styling choices in this kit were refined across five rounds of real feedback on an actual video, and the reasoning behind each decision is written out in a file called SKILL.md rather than just left as final settings. To use it, a person needs video-use already installed, along with ffmpeg and Python, since this kit only adds one extra dependency, the Pillow image library, which video-use's own environment already includes. The kit is meant to be read at the start of a video-use editing session, right after video-use's own instructions, so that brand style, subtitle treatment, and animation layout follow this creator's established defaults instead of being decided from scratch each time. It also includes four standalone helper scripts that can be run directly from the command line to generate individual overlay clips, such as a keyword tag, a corner badge, an opening hook overlay, and a circular picture-in-picture clip. Each script produces a video file meant to be dropped into video-use's project configuration. The repository intentionally does not include any API keys, environment files, or finished videos from the session it was built from, since it is meant to be a reusable style recipe rather than a specific finished project.