WeryAI Music Generator

Generate WeryAI music with the official base skill for vocal-song and music-only workflows. In agent environments, default to an audio-first flow: use wait-music.js to submit and poll until final audio URLs are ready, keeping the wait bounded (10 minutes ceiling). Use submit-music.js plus asynchronous status polling only if the user explicitly asks for async behavior or a task ID. This API is mode-driven rather than model-driven, so the main choice is whether the user wants a VOCAL_SONG or ONLY_MUSIC result.

Example Prompts

• Submit an instrumental generation task and keep checking until you can show me the final audio.
• Make a vocal pop song with warm female vocals and custom lyrics, and give me the audio links.
• Turn this style brief into a piano-only meditation track and preview the request body first.
• Generate a song from this reference audio URL, but use async submit plus status polling instead of waiting in one command.
• Check the status of my WeryAI music generation task and tell me whether the results are ready.

Quick Summary

• Main jobs: music generation, song generation, instrumental generation, task status
• Main modes: ONLY_MUSIC, VOCAL_SONG
• Default mode: VOCAL_SONG
• Default options: genre=none, emotion=none, instrument=none, timbre=male
• Main trust signals: dry-run support, paid-run warning, explicit mode rules, media-source validation with auto upload for local reference_audio