music-generator
wells1137 作
AI & LLMs概要
Generates music from a structured Composition Plan.
説明
このスキルでできること
Takes a structured Composition Plan JSON and generates an audio file using an AI music model. Validates the output against duration, style, and fidelity criteria, then retries generation up to three times on failure.
Built-in quality validation and automatic retries reduce manual review cycles compared to calling a music API directly.
使いどころ
- Generating background music for a video after designing the composition
- Producing instrumental tracks with precise duration matching
- Automating audio output in a content production pipeline
- Retrying music generation when initial output fails quality checks
- Validating audio fidelity before delivery in a music workflow
ワークフロー例
User asks: Generate the final audio file from my composition plan
- 1
Load the Composition Plan JSON file
- 2
Configure generation parameters: duration_ms, output format (WAV/MP3), and instrumental flag
- 3
Call the music generation model API with the plan and parameters
- 4
Validate output for duration match (±0.5s), style consistency, and audio fidelity
- 5
Return the validated file or retry up to 3 times on quality failure
A validated WAV or MP3 audio file matching the Composition Plan specifications
インストール
clawhub install wells1137/music-generatorCLIコマンドを実行するか、Claude Code、Cursor、Codexなどお使いのコーディングエージェントにプロンプトを貼り付けてください。