Six Claude Code / Codex / OpenCode skills that turn source videos into a narrated Chinese recap: scene detection, ASR, VLM scene understanding, script writing, TTS voiceover, and ffmpeg assembly, with optional export to an editable JianYing/CapCut draft. Runs locally with just Python and ffmpeg, calling the Xiaomi MiMo API for ASR/VLM/TTS.
STACK
FFmpegPythonTTS
“Claude Code / Codex skills that turn a video into a Chinese narration recap (视频解说): scene detection, ASR, VLM, script, TTS, ffmpeg assembly, optional editable JianYing / CapCut draft export (剪映草稿导出). Local ffmpeg + one MiMo key, no GPU. | 用 Claude Code skills 把视频做成中文解说成片,可选一键导出可编辑剪映草稿。”
agent-skillsai-agentasrcapcutclaude-codeclaude-code-pluginclaude-skillscodexffmpegjianyingllmmimo
About Editing
Tools that work on footage you already have: cutting, captions, B-roll, silence removal and recaps, often driving FFmpeg or an NLE.