Video Recap Skills
Six skills that let the agent turn video into a short Chinese narration recap, locally via ffmpeg
- 622 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- Runs ffmpeg and scripts, reads and writes video files on your machine
- Sends frames and audio to the external MiMo service with an API key
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/video-recap-skillsDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Runs ffmpeg and scripts, reads and writes video files on your machine
- Sends frames and audio to the external MiMo service with an API key
This is third-party code. Review the repository files before installing.
What it does
The skill installs six linked skills that turn one or more videos into a short recap with Chinese narration. The agent analyzes picture and speech, plans the story and audio, cuts the footage, writes the script, records voiceover and mixes tracks with subtitles. Heavy speech, vision and text-to-speech models run through a Xiaomi MiMo key, while locally only ffmpeg is needed, with no GPU and no model downloads. The multitrack timeline can be exported to an editable CapCut or Jianying draft for manual finishing.
Who it is for. For short-video creators and editors who make Chinese-language recaps of films and shows.
Good fit when
- You want a short recap with narration cut from a long video
- You want the agent to handle editing, script and voiceover instead of doing it by hand
- You need an editable CapCut or Jianying draft for final tweaks
Not a fit when
- You need a recap in a language other than Chinese
- You have no Xiaomi MiMo key or cannot run ffmpeg locally
Example request
Cut these episodes into a 60-second recap with narration and subtitlesLimitations
The narration and voiceover target Chinese. It needs Python 3.10 or newer, ffmpeg with libass in PATH and a Xiaomi MiMo key for speech, vision and synthesis; MiMo is a paid pay-as-you-go service. Voiceover can switch to Fish Audio, but speech and vision still go through MiMo. Video processing runs on your own machine.
How to disable. Remove the video-recap-skills plugin via /plugin or delete the skill folders from the agent's skills directory.
FAQ
Do I need a GPU?
No. Locally it uses only ffmpeg and the Python standard library; heavy models run through the MiMo API.
Can I finish the edit afterwards?
Yes. The timeline exports to a CapCut or Jianying draft where you can edit the original, narration, music, subtitles and overlays.
Discussion
No questions or reviews yet.
Related
SkillsAn agent edits short talking-head videos: transcript, subtitles, B-roll and export for Xiaohongshu, Douyin and WeChat Channels
ChatGPT short-video editing skills
ChatGPT 剪短影音 Skills
Two agent skills: set up the editing environment and build a vertical short video in eight verified steps
A skill that turns a script into a finished video with voice-over, subtitles, a cover and a publish text, frames drawn by code.
A skill that turns a topic into a narrated 4K video: research, script, local TTS, Remotion composition and an FFmpeg mix
A skill that turns an idea, a link or a screen recording into finished video: HeyGen avatar, b-roll, captions and quality checks
A Claude Code skill that downloads a YouTube video, splits it into semantic chapters, cuts chosen clips, translates subtitles and burns bilingual ones in
README in short
The README describes six skills that install into Claude Code, Codex CLI, OpenCode or OpenClaw and turn video into a narrated recap from a natural-language brief. Speech, vision and synthesis go through a Xiaomi MiMo key, while locally only ffmpeg is needed, with no GPU and no model downloads. The agent first makes creative decisions and writes a story plan, then cuts, scripts, voices and assembles. The finished multitrack timeline can export to a Jianying draft for finishing. Every step leaves machine-readable files to check. MIT licensed.
README badge
Original SKILL.md
--- name: video-recap description: > 从输入视频生成中文解说成片或原声剧情短片。用户提供 .mp4 / .mov / .mkv / .webm，并要求剪辑、添加旁白、 配音、总结、短剧/电视剧/电影/纪录片/科普解说时使用。负责编排 video-* 技能链：视频理解 → Agent 制定故事与视听方案 → 剪辑 → 配音 → 合成。触发词：视频解说、视频旁白、生成解说、 视频 recap、video recap、voiceover、narration、auto-dub、recap。 --- ## 1. 定位与流程 本技能是五个独立技能的轻量编排器。各技能只通过 `work_dir` 中的 JSON / MP4 产物通信，不共享代码： video-understanding → Agent 按 video-script 制定方案并写稿 → [video-cut] → video-voiceover → video-assemble 流程支持断点续跑：写好 `narration.json` 后重复同一条命令即可继续。