Individual skill

music-mv

用代码给一首歌做一支音乐 MV（不用视频生成模型）：先把歌听懂（歌词时间轴、节拍、情绪起落、别人怎么理解这首歌），再定画风、画角色、写剧本和分镜、先出草图再做整片，最后照清单看片修改；也能先拆一支参考视频的风格再照着做。画面用 Canvas / p5.js / p5.brush 一帧一帧画，HyperFrames 合成歌词和歌。用户说「给这首歌做个 MV」「做一支音乐 MV / 歌词 MV / 动画 MV」「学这个视频的风格做一支 MV」，或 make a music video / lyric video for this song 时使用。不限模型，能读写文件、跑命令的 agent 都能照做。

SKILL.md

Read pinned source

Source commit 0088b1422f7f82cdf51b456436a6313127a26d6b

What was checked

Matches the observed repository HEAD: Yes
Scanner rules are current: No
Repository scan complete: Yes
Skill scan complete: Yes
Observed behavior: Not checked yet
Last checked: Nov 6, 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): 116
Repository last updated: Oct 20, 2026
License: MIT

Observed agent integration: No specific integration evidence. Standard SKILL.md installation does not imply verified compatibility.

agent-skills Repository topic
claude-code Repository topic
creative-coding Repository topic
hyperframes Repository topic
lyric-video Repository topic
music-video Repository topic
p5js 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

.gitignoreasset · 37 bytes
LICENSEasset · 1,063 bytes
NOTICE.mddoc · 2,027 bytes
README.mddoc · 7,769 bytes
SKILL.mdcore · 9,663 bytes
docs/mv-earth.gifasset · 4,317,240 bytes
docs/mv-fireworks.gifasset · 4,405,463 bytes
docs/mv-seasons.gifasset · 3,482,746 bytes
docs/mv-sky.gifasset · 5,464,801 bytes
docs/style-collage.gifasset · 3,891,243 bytes
docs/style-crayon.gifasset · 4,481,425 bytes
docs/style-diorama.gifasset · 3,860,819 bytes
docs/style-pixel.gifasset · 301,877 bytes
examples/collage/scene.htmlasset · 63,734 bytes
examples/crayon/scene.htmlasset · 52,251 bytes
examples/diorama/scene.htmlasset · 66,201 bytes
examples/pixel/scene.htmlasset · 43,389 bytes
references/character.mddoc · 2,303 bytes
references/learn-style.mddoc · 2,607 bytes
references/render.mddoc · 9,368 bytes
references/review.mddoc · 4,063 bytes
references/storyboard.mddoc · 3,076 bytes
references/styles.mddoc · 6,499 bytes
references/transitions.mddoc · 2,762 bytes
requirements.txtdoc · 21 bytes
scripts/_common.pyruns · 2,119 bytes
scripts/analyze_song.pyruns · 7,926 bytes
scripts/check_env.pyruns · 2,355 bytes
scripts/contact_sheet.pyruns · 7,228 bytes
scripts/finalize_audio.pyruns · 3,073 bytes
scripts/render_frames.mjsruns · 8,831 bytes
scripts/srt2js.pyruns · 1,451 bytes
scripts/video_rhythm.pyruns · 6,352 bytes
templates/frames/scene.htmlasset · 2,461 bytes
templates/hf-canvas/index.htmlasset · 4,535 bytes
templates/hf-overlay/index.htmlasset · 1,898 bytes
templates/lyrics-layer.jsruns · 2,148 bytes
教程.mddoc · 12,099 bytes

Read excerpt


# 用代码做一支音乐 MV

**原理**：不生成视频。代码把每一帧画出来（画面是时间 t 的函数），踩着歌的节拍和歌词编排，最后和原曲合成一条视频。下文 `<skill>` 指本文件所在的文件夹。

## 底线（每一步都守）

1. **一切跟着音乐走**：镜头在哪变、哪个画面最安静、最重要的画面放哪，都从歌里来（`song.md`、`song-analysis.json`）
2. **一句歌词，一个画面**：这句唱什么，画面就演什么。没听过这首歌的人只看画面，也能说出发生了什么
3. **关键处停下来等用户点头**：歌词、画风、角色设定图、分镜、草图。用户没确认，不往下做
4. **画面只由 t 决定**：同一个 t 永远画出同一帧，随机数用固定种子
5. **一个镜头一个画布**：背景、角色、道具画在同一个画布上、跟同一个镜头动，不把单独渲的角色视频叠在背景上
6. **你看不了视频**：节奏和动静靠脚本量（`video_rhythm.py`），画面靠接触表看（`contact_sheet.py`）。每版都跑，不凭感觉说「很流畅」
7. **歌词有版权**：不在对话、命令里整段打出歌词（会被内容过滤拦），脚本读文件，回复里只说第几句
8. **歌词以外的字不自己加**：片名、章节角标、印章徽章，用户要了再加