minimax-music-gen SKILL.md skillA
minimax-music-gen from mini-yifan/open-source-video-gen-skill is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add mini-yifan/open-source-video-gen-skill/minimax-music-gen@v3.0Pins v3.0 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- mini-yifan/open-source-video-gen-skill
- Path
skills/minimax-music-gen/SKILL.md- Versions recorded
- 4
- Latest version
- v3.0, committed Sep 11, 2026
- First version recorded
- Aug 30, 2026
- Size
- 7,854 bytes
- Repository stars
- 40
- License
- a license file GitHub does not identify
用 AutoDL GPU 实例上的 MiniMax Music 3（ComfyUI）生成音乐：人声演唱的完整歌曲 （自定义歌词、女声/男声、任意曲风）或纯音乐 BGM（instrumental）。Use when user wants to generate music, songs, or audio tracks — 生成音乐、写歌、把歌词变成歌曲、生成背景音乐/ 配乐/纯音乐、来首歌、MiniMax Music、BGM、伴奏、情歌 all trigger; match equivalent phrases in any language. Do NOT use for music playback of existing files, music theory questions, or recommendations without generation.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# Skill: minimax-music-gen
# MiniMax Music 3 × AutoDL 实例 音乐生成
在 AutoDL 应用实例的 ComfyUI 里跑 **MiniMax Music 3**（DiT + 专属 text encoder + DAV VAE），
一键完成 **开机 → 提交 → 轮询 → scp 下载 → ffprobe 校验 → 关机**。单首曲目全程约 2~4 分钟。
- 开关机/SSH 通道来自 [`autodl-app-instance`](../autodl-app-instance/SKILL.md) 与
minimax-h3 技能的 `connect_server.py`，本技能只管音乐生成。
- 视频生成走 [`minimax-h3`](../minimax-h3/SKILL.md)，两者共用同一实例批次时遵循
autodl-app-instance 的批次契约：**整批只开一次机、全部完成后只关一次机**。
## 前置条件
| 依赖 | 说明 |
|---|---|
| `AUTODL_TOKEN` | autodl.com → 设置 → 开发者 Token。脚本自动读环境变量或 `~/.config/autodl.env`，缺 Token 时向用户转述脚本的申请指引 |
| 实例 | MINIMAX-H3 应用镜像自带全部音乐模型。已验证可用：`pro-7880531ea6b3`（RTX 5090，2.88 元/小时）。**账号下有多台同名 MINIMAX-H3 实例，必须用 `--uuid` 或 `AUTODL_INSTANCE_UUID` 显式指定**；都不给时脚本会列出候选让你选 |
| 本机工具 | `expect`（macOS 自带，scp/SSH 密码通道必需）、可选 `ffprobe`（时长校验）、可选 `afplay`/`mpv`（播放） |
实例上的模型（缺失说明镜像不对，不要现下载大文件，直接换实例）：
`diffusion_models/minimax_music3/`（DiT 三精度）、`text_encoders/minimax_music3_text_encoder_bf16.safetensors`、
`vae/minimax_music3_dav.safetensors`。
## 生成一条音乐（核心命令）
```bash
python3 "${ZCODE_HOME:-$HOME/.zcode}/skills/minimax-music-gen/scripts/generate_music.py" \
--uuid pro-7880531ea6b3 \
…Read the whole file at its exact version.
How to install
mdr add mini-yifan/open-source-video-gen-skill/minimax-music-gen@sha256:442d2eb1b5e0fbb4The line at the top of this page pins the label v3.0, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_q3afr6ui6nzd5ydm)
3 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 4 versions of this file, between Aug 30, 2026 and Sep 11, 2026, about one every 4 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| v3.0 latest | 2026-09-11 | 3d58d1d | 7,854 B | A | view · diff |
| v3.0 | 2026-09-01 | d62c6b5 | 7,851 B | A | view · diff |
| v2.0 | 2026-08-30 | cee067c | 15,409 B | A | view · diff |
| v2.0 | 2026-08-30 | 8cf2ae7 | 14,134 B | A | view |
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (7854 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
mini-yifan/open-source-video-gen-skill · 40 stars · license a license file GitHub does not identify · pushed 2026-09-11 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_q3afr6ui6nzd5ydm GET https://markdownregistry.com/api/v1/resolve?ref=mini-yifan/open-source-video-gen-skill/minimax-music-gen GET https://markdownregistry.com/api/v1/blob/442d2eb1b5e0fbb432e5ec0e062b939353a397d9abbcb4370914d061a33f915f
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.