Video Production

Structured pipeline for narrated videos: job workspace, TTS voiceover, Manim/ffmpeg render, QC, and archival.

Autor: Aaryan-Kapoor · Aaryan-Kapoor/video-production-skill

Testowano · Działa ★ 9.2/10

Video Production — Structured pipeline for narrated videos: job workspace, TTS voiceover, Manim/ffmpeg render, QC, and archival.

Co robi ten skill

Drives an end-to-end narrated video workflow -- research/source extraction, per-segment narration script, Kokoro (or other) TTS, Manim/ffmpeg visuals synced to voiceover timing, QC frame checks, and archival backups -- inside a durable per-job workspace with a manifest.json. Triggers when the user asks to make, create, generate, render, or produce a video, explainer, paper walkthrough, product demo, or narrated presentation.

Raport z testu

Ran the real pipeline end to end in a throwaway VIDEO_PRODUCTION_ROOT: init_video_job.py created the job/manifest/symlink exactly as documented, synth_kokoro.py --check failed cleanly with the documented install-fallback message (Kokoro not installed), and the ffmpeg mux + ffprobe verify + archive_video.py steps all worked verbatim from the SKILL.md commands, copying the final MP4 into both exports/ and a dated backups/ folder.

Testowano: 2026-07-15 · Claude Code 2.x (agent harness)

Instalacja

git clone https://github.com/Aaryan-Kapoor/video-production-skill
mkdir -p ~/.claude/skills
cp -r video-production-skill/skills/video-production ~/.claude/skills/video-production

Komendy i przykładowe prompty

/video-productionStructured pipeline for narrated videos: job workspace, TTS voiceover, Manim/ffmpeg render, QC, and archival.

Skille uruchamiają się na zwykłe polecenia — bez komend do zapamiętania. Po instalacji aktywują go prompty takie jak te (po angielsku):

Generate a narrated walkthrough video explaining this research paper
Produce a short product demo video with voiceover and synced visuals
Render an explainer video from this script with TTS narration please

Pełny raport →