Claw Score
openclaw/openclaw
Audit or refresh OpenClaw maturity scorecard docs from root taxonomy, maturity scores, and QA evidence artifacts without using maintainer discrawl data or committed inventory reports.
Compiling ARR-SPEC into symbolic notation (符号乐谱输出) - ABC, MusicXML and MIDI from one emitter.
MITAuto-check passed
Install Mc Symbolic Score
$ npx skills add jtydhr88/music-composition-skills --skill mc-symbolic-score -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install jtydhr88/music-composition-skills mc-symbolic-score --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/jtydhr88/music-composition-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/plugins/music-composition/skills/mc-symbolic-score .claude/skills/mc-symbolic-score && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "mc-symbolic-score" agent skill from https://github.com/jtydhr88/music-composition-skills/tree/main/plugins/music-composition/skills/mc-symbolic-score into .claude/skills/mc-symbolic-score/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "mc-symbolic-score", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/jtydhr88/music-composition-skills/tree/main/plugins/music-composition/skills/mc-symbolic-scoreType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add jtydhr88/music-composition-skills --skill mc-symbolic-score -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install jtydhr88/music-composition-skills mc-symbolic-score --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/jtydhr88/music-composition-skills.git skills-src && mkdir -p .agents/skills && cp -r skills-src/plugins/music-composition/skills/mc-symbolic-score .agents/skills/mc-symbolic-score && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "mc-symbolic-score" agent skill from https://github.com/jtydhr88/music-composition-skills/tree/main/plugins/music-composition/skills/mc-symbolic-score into .agents/skills/mc-symbolic-score/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "mc-symbolic-score", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add jtydhr88/music-composition-skills --skill mc-symbolic-score -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install jtydhr88/music-composition-skills mc-symbolic-score --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/jtydhr88/music-composition-skills.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/plugins/music-composition/skills/mc-symbolic-score .cursor/skills/mc-symbolic-score && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "mc-symbolic-score" agent skill from https://github.com/jtydhr88/music-composition-skills/tree/main/plugins/music-composition/skills/mc-symbolic-score into .cursor/skills/mc-symbolic-score/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "mc-symbolic-score", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/jtydhr88/music-composition-skills.git --path plugins/music-composition/skills/mc-symbolic-score--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add jtydhr88/music-composition-skills --skill mc-symbolic-score -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install jtydhr88/music-composition-skills mc-symbolic-score --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/jtydhr88/music-composition-skills.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/plugins/music-composition/skills/mc-symbolic-score .gemini/skills/mc-symbolic-score && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "mc-symbolic-score" agent skill from https://github.com/jtydhr88/music-composition-skills/tree/main/plugins/music-composition/skills/mc-symbolic-score into .gemini/skills/mc-symbolic-score/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "mc-symbolic-score", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install jtydhr88/music-composition-skills mc-symbolic-scoreInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add jtydhr88/music-composition-skills --skill mc-symbolic-score -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/jtydhr88/music-composition-skills.git skills-src && mkdir -p .github/skills && cp -r skills-src/plugins/music-composition/skills/mc-symbolic-score .github/skills/mc-symbolic-score && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "mc-symbolic-score" agent skill from https://github.com/jtydhr88/music-composition-skills/tree/main/plugins/music-composition/skills/mc-symbolic-score into .github/skills/mc-symbolic-score/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "mc-symbolic-score", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add jtydhr88/music-composition-skills --skill mc-symbolic-score -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install jtydhr88/music-composition-skills mc-symbolic-score --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/jtydhr88/music-composition-skills.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/plugins/music-composition/skills/mc-symbolic-score .opencode/skills/mc-symbolic-score && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "mc-symbolic-score" agent skill from https://github.com/jtydhr88/music-composition-skills/tree/main/plugins/music-composition/skills/mc-symbolic-score into .opencode/skills/mc-symbolic-score/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "mc-symbolic-score", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
mc-symbolic-score- GitHub stars
- 154
- Token cost
- ~2.5k tokens
- SKILL.md length
- 816 words
- Files
- 2
- Skills in repo
- 29
- Repo updated
- First seen
- Licence
- MIT
At a glance
Compiling ARR-SPEC into symbolic notation (符号乐谱输出) - ABC, MusicXML and MIDI from one emitter.
- Works in 10 steps: 动笔前必填 → 为什么要走符号路径 → ABC 的映射 → …
- Producing ABC for YuE2
- SKILL.md covers 按任务读哪几节, 边界, 1. 动笔前必填 and 2. 为什么要走符号路径, plus 10 more sections
- Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
What it does
Mc Symbolic Score is an agent skill from jtydhr88/music-composition-skills. Compiling ARR-SPEC into symbolic notation (符号乐谱输出) - ABC, MusicXML and MIDI from one emitter. Why the symbolic path matters (it is the only fully reproducible output and the only way YuE2 executes rather than guesses), the field-by-field mapping from spec to ABC headers and body, roman numerals to letter chords, multi-voice and per-part output, what each target format can and cannot carry, MIDI CC and velocity for expression, and the verification loop that catches a bad emit before it reaches a synth. Use when…
Its SKILL.md is about 2.5k tokens, which your agent loads only when the skill is triggered. The skill folder holds 1 other file (for example `reference.md`).
The repository describes itself as: Professional agent skills for composing and arranging popular music. The licence is MIT.
When your agent uses it
- Producing ABC for YuE2
- MusicXML for a DAW
- A deterministic reference render is needed for A/B
- An LLM should write notation directly
Example prompts
- “/mc-symbolic-score”
Workflow steps
10 steps, taken from the step headings in SKILL.md.
What it can do on your machine
Read from SKILL.md and the folder at commit 7adca0c. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
No scripts in the folder and no shell commands in SKILL.md.
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Mc Symbolic Score loads about 2.5k tokens when it runs. Until then it costs about 186 tokens; SKILL.md has 816 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
The full file from jtydhr88/music-composition-skills at commit 7adca0c, republished under its MIT licence (© jtydhr88). 816 words, ~2,519 tokens.
Download SKILL.mdSave it as
.claude/skills/mc-symbolic-score/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.- name
- mc-symbolic-score
- description
- Compiling ARR-SPEC into symbolic notation (符号乐谱输出) - ABC, MusicXML and MIDI from one emitter. Why the symbolic path matters (it is the only fully reproducible output and the only way YuE2 executes rather than guesses), the field-by-field mapping from spec to ABC headers and body, roman numerals to letter chords, multi-voice and per-part output, what each target format can and cannot carry, MIDI CC and velocity for expression, and the verification loop that catches a bad emit before it reaches a synth. Use when producing ABC for YuE2, when exporting MIDI or MusicXML for a DAW, when a deterministic reference render is needed for A/B, or when an LLM should write notation directly. 符号乐谱、ABC、MusicXML、MIDI、打ち込み、直出乐谱、确定性渲染。
符号乐谱输出（Symbolic Score）
一个发射器，四个目标格式，两个用途。
ARR-SPEC（+ LYR-SPEC 的歌词）
↓ mc-symbolic-score
├─→ ABC → YuE2 的 abc= 输入（★ 让规格被"执行"而不是"提示"）；大模型的母语
├─→ MusicXML → 记谱软件 / 交接给乐手（领奏谱：旋律、和弦、歌词、段落、力度、换气）
├─→ MIDI → DAW ＋ 音源，确定性渲染
└─→ ChordPro → 吉他手、排练（段落、和弦、歌词）为什么是同一个发射器：ABC ↔ MusicXML ↔ MIDI 可互转，ChordPro 是前三者的子集， 四者承载的是同一套信息（音高、时值、小节、调号、拍号、力度、和弦、歌词）。 分成四个 skill 会让同一份规格产生四份不一致的乐谱。 规格单可执行字段的拼写就是按这四种格式定的（schema §1.3.1、§4），发射器不需要翻译符号。
按任务读哪几节
| 任务 | 读 |
|---|---|
| 给 YuE2 供谱 | §3 ABC ＋ §7 |
| 出 MIDI 给 DAW | §5 |
| 出 MusicXML 交给乐手 | §4 |
| 要一个 AB 的确定性基准 | §2.2 |
| 级数转字母 | §3.3 |
| 表情/力度怎么写进 MIDI | §5.2 |
| 发射完怎么验 | §8 |
边界
| 不归这里 | 归哪 |
|---|---|
| 提示词、段落标签、后端选择 | mc-render-compile |
| 旋律/和声/对位的内容本身 | L1 各 skill |
| 用什么音源、采样库选型、keyswitch | 出库 → 制作工具库（不在本包）；MIDI CC 在配器上的含义归 mc-orchestration reference §5 |
| MIDI CC 在配器上的含义 | mc-orchestration reference §5 |
| 量化、velocity 的律动设计 | mc-rhythm-section |
| 记谱法的完整规范（连音线、装饰音写法） | reference.md |
1. 动笔前必填
| 必填 | 问法 |
|---|---|
| ① 目标格式 | ABC / MusicXML / MIDI / ChordPro。能承载的不一样（§6）；meta.deliverable 里列了哪些就发哪些 |
| ② 声部数 | 只发主旋律+和弦？还是全谱？ |
| ③ 谁来读 | YuE2 / DAW / 人。决定详细程度 |
| ④ 这次是不是做基准 | 是的话所有随机性都要关掉 |
2. 为什么要走符号路径
2.1 它是唯一能让规格被"执行"的路
其余后端都只能"提示"。YuE2 的合成步骤直接吃 ABC——
form[].harmony_letters / meta.key / meta.meter / meta.tempo
编译成 ABC 之后，它们就不再是建议了。
把模型自己的符号规划那一步整个旁路掉，把我们的 ABC 直接送进合成步骤。 支持 YuE2 的载体都能这么接（见
mc-render-compile§8.2）；怎么接线是用户环境的事。
2.2 它是唯一完全可复现的输出
同一份规格渲染出来永远一样。
这让它成为 AB 实验的控制变量基准： 想知道"改这一个字段到底带来多大差别"，用符号路径渲染两版，差别 100% 来自那个字段。 生成后端做不到这一点（即使固定种子，改提示词也会改变别的东西）。
2.3 日式打ち込み 工作流本来就是这个形态
LLM 直出 MIDI/MusicXML，进 DAW，配音源。 ★ 这条路没有人声，人声要另接（或走生成后端做人声、符号路径做伴奏）。
3. ABC 的映射
3.1 头部（Header）
| ABC 字段 | 来自 | 例 |
|---|---|---|
X: | 固定 1 | X:1 |
T: | meta.title | T:Night Drive |
M: | meta.meter | M:4/4 |
L: | 默认单位时值（通常 1/8 或 1/16） | L:1/8 |
Q: | meta.tempo | Q:1/4=112 |
K: | meta.key | K:D |
★ tempo_map 有多项时：Q: 取首项，后续变速写成小节内的 [Q:...] 内联域。
3.2 主体
| 内容 | 来自 | 写法 |
|---|---|---|
| 小节线 | form[].bars / start_bar | ` |
| 段落标记 | form[].id + name | ★ 写成注释 % 或 [P:A1] 分段标记，见 §7.2 |
| 和弦 | form[].harmony_letters | "F#m7" 放在音符前的引号里 |
| 旋律 | L1 写的旋律 | 标准 ABC 音高记号 |
| 反复 | 段落重复 | ` |
3.3 ★ 级数 → 字母
ARR-SPEC 里 harmony 是级数、harmony_letters 是字母，两者双记。
发射 ABC 时用字母，但必须校验两者一致。
校验动作：把 harmony_letters 按 meta.key 反推成级数，
与 harmony 逐和弦比对。对不上就是规格里有笔误——
这是 ARR-SPEC 双记号约定的实际用处，不只是为了可读。
移调表见 mc-progressions reference.md §8。
3.4 ★ 反复记号的陷阱
ABC 的 |: :| 会让两遍完全一样。
但本库的 lint #16 要求副歌逐遍有变化——
所以多数情况下不要用反复记号，要把每一遍展开写。
★ 例外：确实完全一样的段落（比如纯器乐的 vamp）可以用反复。 但副歌永远展开写。
4. MusicXML
用途：交给记谱软件、交给乐手。
它比 ABC 多能承载的：
- 完整的谱表布局、声部分配
- 演奏法记号（弓法、连音线、articulation）
- 力度记号（p / mf / f 及渐变）
- 歌词（逐音节对齐）
它比 MIDI 多能承载的：记谱意图（同一个音高时值，是升记号还是降记号、是三连音还是附点）。
★ 什么时候必须用它：mc-orchestration 的产出要交给真人演奏时。
ABC 和 MIDI 都表达不了弓法和分谱。
5. MIDI
5.1 基本映射
| 内容 | 来自 |
|---|---|
| 音符 | 旋律与和声 |
| 轨道/通道 | arrangement.roster[]，一件一轨 |
| 音源映射 | roster[].midi（{program, bank, channel}，GM 号，鼓组 channel 10）；render.midi.instrument_map 只放 VST 名等备注 |
| 速度 | meta.tempo / tempo_map |
5.2 ★ 表情：velocity 与 CC
MIDI 相对 ABC 的最大优势是能写连续控制。
| 来自 ARR-SPEC | 写成 |
|---|---|
arrangement.energy_curve | CC11（表情）曲线 ＋ velocity 层次 |
vocal.dynamics_by_section | 人声轨的 CC11（若有人声音源） |
groove.push_pull | ★ 音符位置的实际偏移（这是唯一能精确执行 push_pull 的路径） |
groove.quantize_policy | 逐轨的量化程度 |
roster[].synth_params | 音源参数或 CC74（亮度）等 |
★ CC7 与 CC11 的分工不能混（→ mc-orchestration reference §5.1）：
- CC7 定这一轨在配器里的位置
- CC11 定这一句怎么呼吸
混用的结果是"推了推子但乐句还是死的"。
5.3 ★ 符号路径上 push_pull 是 exact
backends.yaml 里 symbolic 后端的 groove 相关项是 exact——
因为音符位置可以精确写。
这是验证"push_pull 到底有没有用"的唯一可靠手段：
同一份谱，一版对齐网格、一版按 push_pull 偏移，直接对比。
6. 四种格式能承载什么
| ABC | MusicXML | MIDI | ChordPro | |
|---|---|---|---|---|
| 音高/时值/小节 | ✅ | ✅ | ✅ | ❌ 只有段落 |
| 调号/拍号/速度 | ✅ | ✅ | ✅ | ✅ 头部指令 |
| 和弦符号 | ✅ | ✅ | ⚠ 只能作为文本事件 | ✅ |
| 歌词 | ✅ 基本 | ✅ 完整 | ⚠ 只能作为文本事件 | ✅ |
| 记谱意图（升/降、三连音写法） | 部分 | ✅ | ❌ | ❌ |
| 演奏法记号（弓法、articulation） | ❌ | ✅ | ⚠ 只能靠 CC/keyswitch 近似 | ❌ |
| 连续控制（CC 曲线） | ❌ | ❌ | ✅ | ❌ |
| 微时间偏移（push_pull） | ❌ | ❌ | ✅ | ❌ |
| YuE2 直接吃 | ✅ | ❌ | ❌ | ❌ |
★ 结论：
- 要 YuE2 执行 → ABC
- 要交给人 → MusicXML
- 要表情与律动 → MIDI
- 要给吉他手和排练 → ChordPro
- 几个都要就几个都发，但从同一份规格发，不要各写各的
6b. YuE2 方言：实测出来的四条（2026-09-17）
同一份从 ARR-SPEC 编译的谱，三种写法送进 YuE2 的 abc 口，结果差得远：L:1/16 单声部出 2:18（谱面 4:00，模型提前停），删掉段落注释更短（1:58），改成 L:1/32 跑满 3:56。所以给 YuE2 的 ABC 照它自己写谱的样子来：
L:1/32：所有时值按三十二分音符计，四分音符写 8、附点二分写 24、整小节休止z32。用L:1/16它会把谱当成一半长，提前结束。 ⚠ 但只改这一条不够：同一份谱改成L:1/32后时长跑满，whisper 却一个字都识别不出（直出版 76%，模型自写谱 65%，这版 2%）。原因在旋律本身：每个字的时值太长、音太少、音区比模型自写的谱低一个八度。给 YuE2 的人声线要像它自己写的那样：小写音名区（c 到 a 一带）、每小节四到六个音、单字很少超过一拍。实测：同一份谱整体升八度到小写音名区，whisper 识别率从 2% 到 47%；再加上 w: 歌词行到 56%（直出是 76%）。这就是第 5、6 条的来历：带
w:歌词行（见第 6 条）。声部头写
V: Vocal clef=treble name="Vocal Melody" snm="Vocal"，段落用% intro／% verse／% pre-chorus／% chorus／% bridge／% outro注释；它自写的谱还有一条V: Ins声部放器乐句，单声部也能跑满。标题行
T:可空；调用K:D这类标准写法，模型自写时倾向关系小调（同一首它写成K:Dm），不影响。人声线写在小写音名区（ABC 的 c 到 a，即 C5 到 A5 一带），大写区它会唱成含混的元音；这与真人音域无关，是模型的习惯。
带
w:歌词行（每个音符一个字，字数与音符数逐行相等）比只在 lyrics 口给词识别率高；w: 行照带，lyrics 口也照给。待做：旋律密度。编译器目前给每个字配长音（附点二分、全音符），模型自写谱每小节四到六个音、单字很少超过一拍；识别率剩下的差距多半在这里。
ABC 表达不了的规格字段（减法事件、残响切断、逐件进退场）只能靠 style 文本，见 mc-render-compile。
7. 给 YuE2 供谱的实操
Show full SKILL.md (327 more words)Show less
7.1 接法
[我们的 ABC] ──→ YuE2GenerateMusic ──→ 音频
↑
（YuE2GenerateABC 被旁路掉）本 skill 只管左边那一格：把 ARR-SPEC 编译成 ABC 文本。
右边这条线为什么要接、接了之后规格为何不再只是"提示"、
用什么载体（官方脚本、ComfyUI 节点）接线，见 mc-render-compile §8.2。
7.2 方言已经摸清
YuE2GenerateMusic 的 abc 输入端吃什么，§6b 已按实测写死：段落用 % 注释，和弦用引号里的标准拼写（"F#m7"），V: Vocal 加可选的 V: Ins 两个声部都能跑。仍不确定的东西只剩旋律密度那一刀（§6b 第 7 条）。
7.3 上游有官方 skill
YuE2 官方仓库自带 skills/yue2-music/（SKILL.md 格式，含"检查音乐不变量"）。
★ 不要重写它。 我们写 ARR-SPEC→ABC 的适配器，模型操作的知识引用它们的。 它的 musical invariants 应该与我们的 20 条自查 对齐或互补——
8. ★ 发射后的验证回路
不要把没验过的乐谱送进合成器。 三步：
| # | 检查 | 怎么做 |
|---|---|---|
| 1 | 语法 | ABC → 用解析器过一遍；MusicXML → schema 校验；MIDI → 读回来看轨数与时长 |
| 2 | 与规格一致 | ★ 小节总数 == Σ form[].bars；调号拍号速度对得上；字母和弦反推级数 == harmony |
| 3 | 听一遍最简渲染 | 用任意音源快速渲染，只听结构对不对，不听音色 |
★ 第 2 步的小节总数校验最值钱—— 发射器最常见的 bug 是段落边界算错，而那个错误在音频里表现为"结构乱了"， 很容易被误判成后端不听话。
这三步用眼睛就能查；仓库内部若做自动化也不随包发布。
实测过的一条路（2026-09）：本 skill 写法的 ABC（含 w: 行、引号和弦、% 段落注释、V: 头）用 music21 解析零警告，能直接转出 MIDI 与 MusicXML，MIDI 配任意 GM 音色就是第 3 步的最简渲染，不经过生成模型，倒字与旋律密度在这一步就能听到。两个注意：music21 不读 w: 行，交给乐手的 MusicXML 要用会读 w: 的转换器（abc2xml、abcm2ps 一类）或按"每个音符领一个音节"自行贴回；引号和弦默认不发声，想听和声垫底要自己把它们展开成一条钢琴声部。
9. 与 ARR-SPEC 的字段对应（四个导出目标）
规格单每个字段都有去向（schema §4）。可执行的字段按下表落到四种格式；本 skill 负责其中的发射规则。
| ARR-SPEC 字段 | ABC | MusicXML | MIDI | ChordPro | 本 skill |
|---|---|---|---|---|---|
meta.title | T: | work-title | 轨名 | {title} | §3.1 |
meta.key / mode | K:（K:Ddor） | key 的 fifths + mode | 调号 meta | {key} | §3.1 |
meta.meter / meter_map | M:，行内变拍 | time，逐小节 | 拍号 meta | {time} | §3.1 |
meta.tempo / tempo_map | Q: | metronome / sound tempo | set_tempo | {tempo} | §3.1 |
form[].id / name | P: 或 % 注释 | rehearsal 记号 + words | marker | {start_of_verse} 等 | §3.2 |
form[].bars / start_bar | 小节线 | measure number | tick | §3.2；§8 第 2 步 | |
form[].harmony_letters | "F#m7" | harmony 的 root / kind / bass，映射见 schema §1.3.1 | 和弦轨或 marker 文本 | [F#m7] | §3.3 |
material.motif_abc / arrangement_hook.abc | 原样嵌入 | 实际音符 | 实际音符 | §3 | |
roster[].id / midi | V: 名 | part-name + midi-instrument | 轨名 + program change | §5.1 | |
roster[].entry / exit、form[].events | 段前 Z 多小节休止 | 多小节休止 | 轨内容范围 | §5.1 | |
energy_curve、vocal.dynamics_by_section | !p! !f! 力度记号 | dynamics（映射见 schema §1.5） | 力度缩放 / cc11 | §5.2 | |
groove.push_pull | 无 | 无 | tick 偏移（★ 唯一 exact 的路径） | §5.3 | |
vocal.breath_points | !breath! | breath-mark | 无 | §4 | |
mix_intent.sound_stage[].pan | 无 | midi-instrument 的 pan | cc10 | §5.1 | |
depth_map | 无 | 无 | cc91 | §5.2 | |
LYR-SPEC lyrics[].lines | w: 行 | lyric，syllabic=single，中文一字一音节 | 无 | 歌词行 | §3 |
render.yue.abc | 本体 | §7 | |||
render.musicxml / render.midi / render.chordpro | 目标开关 | 目标开关 | 目标开关 | §4 §5 |
仅意图的块（intent、material 的文字、mix_intent 与 vocal 的大部分）整块作为文本随文件走：ABC 用 % 注释，MusicXML 用 miscellaneous，MIDI 用第一小节的 text 事件。四个目标各给谁、带什么，见 schema §4。
10. 发射完必过
- 目标格式是按 §6 选的，不是习惯性选 MIDI
- 小节总数 ==
Σ form[].bars - 调号、拍号、速度与
meta一致 - ★ 字母和弦反推级数，与
harmony逐和弦对得上 - 副歌各遍是展开写的，没有用反复记号（lint #16）
- MIDI 里 CC7 与 CC11 分工正确
- 若要验证 push_pull，做了对齐版和偏移版两份
- 送进合成器前跑过 §8 三步
- 给 YuE2 供谱时，用的是已确认的 ABC 方言子集
附：来源
- §8 的三步校验是人工的；其中"小节总数"与"和弦级数反推"两项是纯数据校验，不需要听。
- ⚠ 未确认：YuE2 的 ABC 方言（§7.2）
- 记谱法细则、ABC 语法速查、GM 音色表在 reference.md。
© jtydhr88, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
SKILL.md and 1 other file in plugins/music-composition/skills/mc-symbolic-score of jtydhr88/music-composition-skills.
- SKILL.md
- reference.md
Open the folder on GitHubat commit 7adca0c
Compare with similar skills
Mc Symbolic Score next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Mc Symbolic Score this skilljtydhr88/music-composition-skills | 154 | — | ~2.5k | Automated safety check: Pass | MIT | |
| Claw Scoreopenclaw/openclaw | 392k | — | ~2.5k | Automated safety check: Pass | MIT | |
| Harness Scoreruvnet/ruflo | 74k | — | ~605 | Automated safety check: Notes | MIT | |
| Score Evalsickn33/agentic-awesome-skills | 47k | 1 repos | ~304 | Automated safety check: Pass | MIT | |
| UI Scoresickn33/agentic-awesome-skills | 47k | 1 repos | ~1.8k | Automated safety check: Pass | MIT | |
| Compile Knowledgesickn33/agentic-awesome-skills | 47k | 1 repos | ~2.3k | Automated safety check: Pass | MIT |
Similar skills
Harness Score
ruvnet/ruflo
5-dimension harness readiness scorecard from metaharness score {path}.
Score Eval
sickn33/agentic-awesome-skills
Imported skill score-eval from upstream source. An agent skill from sickn33/agentic-awesome-skills.
UI Score
sickn33/agentic-awesome-skills
Score a UI file's design quality 0-100 against StyleSeed's design language — per-category breakdown, the worst offenders, and a prioritized fix list.
Compile Knowledge
sickn33/agentic-awesome-skills
Compile durable, non-obvious findings into an interlinked markdown knowledge store — atomic files, [[wiki-links]], a maintained index — so an agent gets smarter across sessions instead of relearning…
Compiler
atopile/atopile
How the atopile compiler builds and links TypeGraphs from .ato (ANTLR front-end → AST → TypeGraph → Linker → DeferredExecutor), plus the key invariants and test entrypoints.
More from jtydhr88/music-composition-skills
All 29 skills in this repoMc AI Tell Audit
jtydhr88/music-composition-skills
The post-generation AI-tell audit (AI 味诊断). An agent skill from jtydhr88/music-composition-skills.
Mc Arrangement Arch
jtydhr88/music-composition-skills
Arrangement architecture (编曲结构学) - the shape of a track over time rather than its notes.
Mc Case Studies
jtydhr88/music-composition-skills
The corpus layer (语料层) - how real recordings are made to testify for or against the textbook rules.
Mc Development
jtydhr88/music-composition-skills
Thematic development (乐思发展层) - what the theme is and what each section does with it.
Mc Form
jtydhr88/music-composition-skills
Song form and section design (曲式与段落) - the terminology, the standard templates, and how to build a setsu backwards from the chorus.
Mc Melody
jtydhr88/music-composition-skills
Melody writing (旋律法) - what a melodic line expresses and how to make it singable.
Questions about Mc Symbolic Score
What does Mc Symbolic Score do?
Compiling ARR-SPEC into symbolic notation (符号乐谱输出) - ABC, MusicXML and MIDI from one emitter. Mc Symbolic Score is an agent skill from jtydhr88/music-composition-skills. Compiling ARR-SPEC into symbolic notation (符号乐谱输出) - ABC, MusicXML and MIDI from one emitter.
When should I use Mc Symbolic Score?
Mc Symbolic Score fits situations like: producing ABC for YuE2; musicXML for a DAW; A deterministic reference render is needed for A/B; an LLM should write notation directly.
How do I install Mc Symbolic Score in Claude Code?
Run `npx skills add jtydhr88/music-composition-skills --skill mc-symbolic-score -a claude-code`. Or copy the skill folder (plugins/music-composition/skills/mc-symbolic-score in jtydhr88/music-composition-skills) into .claude/skills/mc-symbolic-score in your project. Claude Code loads it when a task matches its description.
How do I install Mc Symbolic Score in Codex?
Run `npx skills add jtydhr88/music-composition-skills --skill mc-symbolic-score -a codex`. Or copy the skill folder (plugins/music-composition/skills/mc-symbolic-score in jtydhr88/music-composition-skills) into .agents/skills/mc-symbolic-score in your project. Codex loads it when a task matches its description.
Can I use Mc Symbolic Score in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add jtydhr88/music-composition-skills --skill mc-symbolic-score -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/mc-symbolic-score, .gemini/skills/mc-symbolic-score, .github/skills/mc-symbolic-score and .opencode/skills/mc-symbolic-score in your project.
What does Mc Symbolic Score need to run?
SKILL.md names no scripts, command-line tools or credentials: Mc Symbolic Score is instructions for the agent only.
Does Mc Symbolic Score access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Mc Symbolic Score safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Mc Symbolic Score use?
Mc Symbolic Score is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Mc Symbolic Score use?
About 2.5k tokens (SKILL.md is roughly 10k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Mc Symbolic Score?
Skills that share tags, products or a category with Mc Symbolic Score: Claw Score (openclaw/openclaw, 392k stars), Harness Score (ruvnet/ruflo, 74k stars), Score Eval (sickn33/agentic-awesome-skills, 47k stars) and UI Score (sickn33/agentic-awesome-skills, 47k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Mc Symbolic Score?
jtydhr88 (a GitHub user) maintains it in jtydhr88/music-composition-skills, which has 154 GitHub stars. The repository holds 29 skills in this directory. The repository was last updated on September 22, 2026.
Source: jtydhr88/music-composition-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.