multimodal-audiocraft — Claude Code skill: setup and features
Equips Claude Code with workflows for text-to-music and sound generation using Meta's AudioCraft tools. Helpful when building audio synthesis pipelines.
npx claude-code-templates@latest --skill ai-research/multimodal-audiocraftWhat multimodal-audiocraft does
This skill gives Claude Code workflows for generating music and audio effects. It covers MusicGen and AudioGen. The templates handle text prompts alongside melody conditioning with reference audio. Under the hood, it details T5 embeddings and EnCodec decoding.
Running the code requires torch>=2.0.0 and audiocraft. Large model checkpoints will require a capable GPU. It also includes standard bounds for duration and guidance scales.
What it helps with
- write Python code for text-to-music generation with MusicGen
- synthesize background sound effects using AudioGen
- condition song generation on an input melody track
- configure sampling parameters like temperature and guidance scale
Example prompts for the agent
- Generate a 30-second electronic track prompt using MusicGen
- Write a script to synthesize barking dog sounds with AudioGen
- Show how to pass melody.wav into musicgen-melody with torchaudio
How to install multimodal-audiocraft
1claude-code-templates
npx claude-code-templates@latest --skill ai-research/multimodal-audiocraft2Skills CLI (Vercel)
npx skills add davila7/claude-code-templates --skill multimodal-audiocraft -a claude-code3Manually: put the files in place
~/.claude/skills/multimodal-audiocraft/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/multimodal-audiocraft/
Manually: put the files in place
.agents/skills/multimodal-audiocraft/SKILL.mdFor all projects use ~/.agents/skills/
Manually: put the files in place
.cursor/skills/multimodal-audiocraft/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install davila7/claude-code-templates multimodal-audiocraft2Manually: put the files in place
.github/skills/multimodal-audiocraft/SKILL.mdPlatform CLI
gemini skills install https://github.com/davila7/claude-code-templates --path cli-tool/components/skills/ai-research/multimodal-audiocraftManually: put the files in place
.devin/skills/multimodal-audiocraft/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about multimodal-audiocraft
How do I install multimodal-audiocraft in Claude Code?
Run npx claude-code-templates@latest --skill ai-research/multimodal-audiocraft in your console. You can also place the template manually into ~/.claude/skills/multimodal-audiocraft/SKILL.md.
What does multimodal-audiocraft do?
It provides prompt templates and code examples for MusicGen and AudioGen models. It helps Claude Code generate audio clips and handle melody conditioning.
What dependencies are required to run the generated code?
The environment needs audiocraft, torch>=2.0.0, and transformers>=4.30.0.