Individual skill
video-shots
拉片:把一条成片拆成逐镜头的分析表——每个镜头的时长、景别、类别、运镜、画面。 分工刻在骨子里:**能量的都由代码量**(切点来自 ffmpeg 场景检测,时长是切点相减, 运动量是逐帧差分的中位数),模型只判断它真正该判断的那几件事(景别 / 类别 / 运镜 / 画面 / 节奏), 然后每一条判断都被代码当场对账——**声称推拉摇移却实测几乎不动,门直接拦**。 看片走联系表(每镜起手帧 + 收尾帧各拼一张大图,一屏二十几个镜头,a/b 对照就是运镜), 不是一张张翻。检测漏刀多刀用 recut 补刀并刀,自动重编号重算时长,手改边界过不了门。 产出 shots.json + Markdown 镜头表 + **单页交互式拉片报告**:内嵌播放器(播放时同步高亮镜头、 点镜头跳转)、镜头节奏带、可搜索可筛选可排序的镜头表(列表 / 卡片两种视图、首尾关键帧并排、 点图开大图)、景别类别运镜分布、出场人物、质量门、导出 JSON。单文件零依赖,离线双击能开。 15 道质量门全部由脚本确定性检查。 零依赖、零 API key,只要 node 和 ffmpeg。 Use when asked to 拉片、拆镜头、分析视频镜头、镜头时长、景别、运镜、镜头表、 video shot breakdown、shot list from video。
skills/video-shots/SKILL.md
Source commit df39d5df9d7adc2f82acba5594579d60361a42c7
What was checked
- Matches the observed repository HEAD
- Yes
- Scanner rules are current
- Yes
- Repository scan complete
- No
- Skill scan complete
- No
- Observed behavior
- Network or broad shell
- Last checked
- Nov 10, 2026
These are observations at the recorded commit, not a guarantee of safety. Review the source before running it.
Install this skill
Choose an agent in your Stack or copy a command below. Commands fetch the pinned commit and copy only this skill directory. Existing destinations are refused. Copying a command does not install anything.
An install command is unavailable: current rules, complete repository and skill scans, an eligible behavior level and a permitted license are required.
Repository context
- Repository stars (not skill ratings)
- 1,209
- Repository last updated
- Nov 9, 2026
- License
- Apache-2.0
Observed agent integration: No specific integration evidence. Standard SKILL.md installation does not imply verified compatibility.
- ai-video Repository topic
- claude-skills Repository topic
- ffmpeg Repository topic
- shot-analysis Repository topic
- video-analysis Repository topic
- engineering Derived category
Files and excerpt
File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.
Browse files
README.en.mddoc · 9,868 bytesREADME.mddoc · 8,467 bytesSKILL.mdcore · 13,821 bytesSKILL.mdcore · 13,821 bytesSKILL.mdcore · 13,821 bytesSKILL.mdcore · 13,821 bytesSKILL.mdcore · 13,821 bytesSKILL.mdcore · 13,821 bytesSKILL.mdcore · 13,821 bytesSKILL.mdcore · 13,821 bytesSKILL.mdcore · 13,821 bytesSKILL.mdcore · 13,821 bytesSKILL.mdcore · 13,821 bytesSKILL.mdcore · 13,821 bytesSKILL.mdcore · 13,821 bytesassets/report-mobile.pngasset · 128,870 bytesassets/report.pngasset · 393,735 bytesexamples/demo-shots.jsonasset · 31,686 bytesexamples/demo-track.jsonasset · 3,926 bytesreferences/analysis-pass.mddoc · 5,624 bytesreferences/report-style.mddoc · 5,579 bytesreferences/schema.mddoc · 5,372 bytesreferences/taxonomy.mddoc · 11,939 bytesscripts/report.cssasset · 33,842 bytesscripts/report.jsruns · 20,988 bytesscripts/selftest.mjsruns · 36,055 bytesscripts/video-shots.mjsruns · 77,449 bytes
Read excerpt
## video-shots
给成片**拉片**——把一条片子拆成逐镜头的分析表:**时长、景别、类别、运镜、画面**。
**前提刻在骨子里:镜头边界是量出来的,不是看出来的。** 模型看视频最不可靠的就是报时间——
「这个镜头大概 3 秒」和「2.97 秒」差的不是精度,是这份表能不能用。所以这里划一条死线:
| 谁来定 | 什么 | 怎么定 |
| --- | --- | --- |
| **代码** | 切点、时长、片长、帧率 | ffmpeg 场景检测 + ffprobe,两位小数 |
| **代码** | 每个镜头的实测运动量 | 逐帧差分曲线的区间中位数(两端剔除,避开切点尖峰) |
| **模型** | 景别、类别、运镜、画面、节奏 | 看关键帧判断——**只有这几件事是模型的活** |
| **代码** | 判断对不对 | 15 道质量门,逐条对账 |