Uses ElevenLabs, Diffrythm, and Tencent models via inference.sh CLI to generate everything from 10-second podcast jingles to full songs with vocals. ElevenLabs gives you commercial licensing for up to 10 minutes, while Diffrythm handles quick instrumental generation and Tencent does complete tracks with lyrics. Good for content creators who need background music without dealing with royalties, or game developers building soundtracks. The prompting is straightforward with genre and mood keywords. You'll need to install their CLI first, but then it's just running commands with JSON inputs to get audio files back.
npx skills add https://github.com/inferen-sh/skills --skill ai-music-generation