What it does

This skill enables your AI agent to act as a high-fidelity music producer. By leveraging Google's Lyria 3 models via the Gemini API, it generates everything from 30-second background clips to full 2-3 minute instrumental or vocal tracks from simple text descriptions. It handles the API handshake, directory management, and file serialization so you get a ready-to-play MP3 instantly.

Why use this skill

While standard LLMs can only write lyrics or chord progressions, this skill bridges the gap to actual audio production. It is significantly more efficient than manual API orchestration because it manages environmental dependencies and output handling. Plus, it utilizes the free tier of the Google GenAI SDK, providing professional-grade 48kHz stereo output without per-track costs.

Supported features

• Model Versatility: Supports lyria-3-clip-preview for quick loops and lyria-3-pro-preview for full song structures.
• Deep Control: Fine-tune output using BPM, musical key, specific instruments (e.g., 808 drums, Rhodes piano), and mood descriptors.
• Structured Composition: Use timestamped prompts to define transitions between intros, verses, and choruses.
• High Quality: Produces SynthID-watermarked, crystal-clear 48kHz audio files.