菜单
Analyzes audio files for BPM, musical key, loudness (LUFS), duration, format, and quality metrics. Uses ffprobe for metadata and FFmpeg for loudness measurement.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/AgriciDaniel/claude-music --skill claude-music-analyze命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Music production suite using ACE-Step 1.5 via Python API. Routes /music commands for generation, cover, repaint, compose, analyze, export, enhance, random, and LoRA training. 50+ languages, up to 10-minute tracks, 48kHz stereo.
Starts the local claude-music web dashboard: a minimalist browser app for generating songs with style buttons, live progress, a waveform player, a rated library, downloads, and one-click similar generations. Runs entirely on 127.0.0.1 with no cloud calls.
Creates cover versions and style transfers of existing songs using ACE-Step 1.5. Takes a reference audio file and generates a new version with different style, genre, or vocal characteristics while preserving musical structure.
基于 SOC 职业分类
正在显示 SKILL.md
| name | claude-music-analyze |
| description | Analyzes audio files for BPM, musical key, loudness (LUFS), duration, format, and quality metrics. Uses ffprobe for metadata and FFmpeg for loudness measurement. |
| when_to_use | Use when the user asks to analyze, inspect, or measure an audio file — BPM, key, loudness, LUFS, duration, codec, or general audio info. |
| allowed-tools | ["Bash","Read"] |
# Full metadata
ffprobe -v quiet -print_format json -show_format -show_streams input.flac
# Duration only
ffprobe -v quiet -show_entries format=duration -of csv=p=0 input.flac
# Sample rate, channels, codec
ffprobe -v quiet -show_entries stream=sample_rate,channels,codec_name -of csv=p=0 input.flac
ffmpeg -i input.flac -af loudnorm=I=-14:TP=-1:LRA=11:print_format=json -f null - 2>&1 | \
grep -A 20 '"input_'
Key fields in output:
input_i: Integrated loudness (LUFS)input_tp: True peak (dBTP)input_lra: Loudness range (LU)Using ffmpeg's ebur128 for rhythm analysis, or for accurate BPM:
# If librosa is available in ~/.video-skill/ venv:
source ~/.video-skill/bin/activate
python3 -c "
import librosa
y, sr = librosa.load('input.flac', sr=None)
tempo, _ = librosa.beat.beat_track(y=y, sr=sr)
print(f'BPM: {tempo[0]:.1f}' if hasattr(tempo, '__len__') else f'BPM: {tempo:.1f}')
"
source ~/.video-skill/bin/activate
python3 -c "
import librosa
import numpy as np
y, sr = librosa.load('input.flac', sr=None)
chroma = librosa.feature.chroma_cqt(y=y, sr=sr)
key_names = ['C', 'C#', 'D', 'D#', 'E', 'F', 'F#', 'G', 'G#', 'A', 'A#', 'B']
key_idx = np.argmax(np.mean(chroma, axis=1))
print(f'Estimated key: {key_names[key_idx]}')
"
Combine all analyses into a single report: