Individual skill

yue2-instrumental

用 YuE2 生成纯音乐或制作纯音乐 cover。默认由 YuE2 写谱，自动将 Vocal 音符移到 Ins 后生成；只有用户明确要求 agent 作曲时才由 agent 写谱。支持文字描述、ABC 或音频参考，交付试听、真实 prompt 和转谱检查。

skills/yue2-music/instrumental/SKILL.md

Read pinned source

Source commit 1116bea76032bee5668e5f097294d082fa0b84f0

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 9, 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): 13,266
Repository last updated: Nov 1, 2026
License: Apache-2.0

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

foundation-models Repository topic
music-generation Repository topic
huggingface Repository topic
llama Repository topic
audio-generation Repository topic
voice-cloning Repository topic
llms Repository topic
style-transfers Repository topic
ai Repository topic
deep-learning 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

LICENSEasset · 1,080 bytes
PRIVACY.mddoc · 2,381 bytes
README.mddoc · 2,889 bytes
SKILL.mdcore · 4,291 bytes
SKILL.mdcore · 4,291 bytes
agents/openai.yamlasset · 312 bytes
assets/cover-release.jsonasset · 603 bytes
assets/example-ins.abcasset · 305 bytes
assets/example-original.abcasset · 314 bytes
assets/example-repairs.jsonasset · 772 bytes
assets/example-style.txtdoc · 269 bytes
assets/release.jsonasset · 2,151 bytes
assets/validation.jsonasset · 8,698 bytes
bundle-manifest.jsonasset · 3,768 bytes
references/abc-repair.mddoc · 5,123 bytes
references/cover.mddoc · 2,783 bytes
references/event-schema.mddoc · 1,780 bytes
references/setup.mddoc · 5,128 bytes
requirements-inference.txtdoc · 192 bytes
scripts/abc_tools.pyruns · 15,750 bytes
scripts/common.pyruns · 951 bytes
scripts/compile_score.pyruns · 10,540 bytes
scripts/instrumental.pyruns · 27,768 bytes
scripts/instrumentalize.pyruns · 5,980 bytes
scripts/setup_runtime.pyruns · 5,991 bytes
scripts/test_tools.pyruns · 27,776 bytes
scripts/transcribe_cover.pyruns · 6,300 bytes
scripts/verify_bundle.pyruns · 2,338 bytes

Read excerpt


# YuE2 纯音乐与器乐 cover

用户只需描述音乐或提供参考。你负责环境、转谱、运行和交付；用户不需要懂声部格式或推理参数。

推荐使用 GPT-6 Astra 作为执行本 skill 的 agent；默认作曲与声音生成仍由 YuE2 完成。

## 选择来源

- **文字生成，默认模式**：调用 YuE2 的符号规划器写谱，保存原谱，将 Vocal 旋律移到 Ins 后重新生成音频。不要自己写 events/ABC 替代这一步，也不要套用示例旋律。
- **用户明确要求 agent 作曲**：才由你写谱，使用 `--composer agent`。复杂谱可以写 [events.json](references/event-schema.md)。
- **用户给了 ABC**：以用户曲谱为准，直接转为纯音乐条件，不重新作曲。只有不规范输入才读 [修谱规则](references/abc-repair.md)；保留原件与改动说明。
- **纯音乐 cover**：读 [cover 流程](references/cover.md)。音频先经 SheetSage2 转谱，再做相同的 Vocal → Ins 转换。默认保留旋律、让 YuE2 重新配伴奏；要求保留原和声时加 `--keep-harmony`。这不是直接去人声或伴奏分离。

模型规划失败或曲谱无法解析时保留原始输出，定位问题后至多重试一次。不要自动切换成 agent 作曲；音乐性修改必须符合用户授权。严格时长和音频逐音跟谱不是模型保证。