Agent skill

beat

Stars 163
Forks 31

Install this agent skill to your Project

npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/data/beat

SKILL.md

Beat Generator

16ステップビートシーケンサー。スタイルプロンプトから JSON / MIDI / WAV / MP3 を生成する。

Context (auto-collected)

実行時に以下を取得:

  • run_id: date +%Y%m%d_%H%M%S
  • ffmpeg: which ffmpeg || echo "(no ffmpeg)"

Pipeline

Step 0: Prepare dirs

  • beats/.beatlab/ を作成
  • 出力先: beats/<run_id>/

Step 1: Generate (up to 3 attempts)

最大3回リトライ:

  1. music-reference-agent でスタイルプロンプト → BeatSpec YAML
  2. music-generation-agent で BeatSpec → Beat JSON
  3. JSON を beats/<run_id>/beat.json に保存
  4. バリデーション:
    bash
    uv run python .claude/skills/beatlab-pipeline/scripts/validate.py beats/<run_id>/beat.json --inplace
    
  5. 失敗時は music-generation-agent に修正を依頼してリトライ

Step 2: Update current

bash
cp beats/<run_id>/beat.json .beatlab/current.json

Step 3: Render grid

bash
uv run python .claude/skills/beatlab-pipeline/scripts/render_grid.py beats/<run_id>/beat.json

Step 4: Export files

bash
# MIDI
uv run python .claude/skills/beatlab-pipeline/scripts/export_midi.py beats/<run_id>/beat.json beats/<run_id>/beat.mid

# WAV
uv run python .claude/skills/beatlab-pipeline/scripts/render_wav.py beats/<run_id>/beat.json beats/<run_id>/beat.wav

# MP3 (ffmpeg がある場合)
ffmpeg -y -i beats/<run_id>/beat.wav beats/<run_id>/beat.mp3

Output Format

レスポンスに含める:

  • Summary: bpm, swing, run_id
  • ファイルパス: beat.json / beat.mid / beat.wav (/ beat.mp3)
  • ASCII 16-step grid

Expand your agent's capabilities with these related and highly-rated skills.

Didn't find tool you were looking for?

Be as detailed as possible for better results