This is the captioned video pipeline you reach for when you have raw footage and need polished subtitles in any language Whisper supports. It transcribes, gives you a browser-based review step to fix mishears before rendering (critical for Hebrew), then builds a HyperFrames composition with liquid-glass caption pills and blob backgrounds. The smart bit is the mandatory approval gate: no wasted 20-minute renders on typos. It also handles dual-aspect output from a single 16:9 source, generating both landscape and portrait versions with a blurred-backdrop treatment for vertical. Pairs with the hyperframes stack and sits one tier below yuv-pilot in the orchestration pyramid.
npx -y skills add hoodini/ai-agents-skills --skill video-edit --agent claude-codeInstalls into .claude/skills of the current project.
Select a file.
sickn33/antigravity-awesome-skills
kubesphere/kubesphere
supercent-io/skills-template