jev-skip

Reads YouTube captions, classifies segments as sponsorship, introduction or main content, and marks and skips segments that meet the threshold. It does not work without captions. The README states that its demonstrations and metrics mainly come from recorded samples and should not be treated as complete live validation. Your own TypeSafe key is required. Source code is freely available. Online models or third-party services may charge additional API fees. Sources: https://github.com/valentynkit/jev-skip/blob/main/README.md https://www.jev.store/projects/jev-skip Only public source code or documentation was checked for this listing. The application was not installed, run or tested, and its actual effectiveness was not verified. Project demonstration video:

The extension interface and playback run in the recording, while its classifications are replayed from API responses saved on 18 September 2026 through a local server. This is not a fresh model call. Source video: Corridor Crew, VFX Artists React to BOLLYWOOD Bad & Great CGi 8 (https://www.youtube.com/watch?v=2wuUSEFjr50).