music-creative

name	music-creative
description	Music creative suite: AI music generation (Suno/HeartMuLa), audio analysis (songsee), and songwriting craft.
version	1.0.0
author	Hermes Agent
license	MIT
platforms	["linux","macos","windows"]
metadata	{"hermes":{"tags":["music","songwriting","audio-analysis","AI-music","Suno"]}}

Music Creative Suite

Unified skill for all music-related creative workflows. Three modes: AI music generation, audio feature analysis, and songwriting craft.

---

Section A: AI Music Generation (HeartMuLa + Suno)

Generate music from lyrics + tags using HeartMuLa (local) or Suno API.

HeartMuLa

Local song generation from lyrics and style tags. Use when you need full control over generation without external API dependencies.

Workflow: Provide lyrics + genre/style tags → generate audio → deliver to user.

Suno AI Music

Use the songwriting-and-ai-music skill for Suno AI prompts and workflows.

Key conventions:

• Lyrics should be structured (verse, chorus, bridge)
• Style tags control genre, mood, instrumentation
• Always specify tempo, key, and instrumentation preferences
• Suno generates full songs (~2-4 minutes)

---

Section B: Audio Analysis (songsee)

Extract audio spectrograms and features (mel, chroma, MFCC) via CLI.

Use when: User asks about audio characteristics, wants to analyze a song's structure, or needs feature extraction for ML/audio ML tasks.

Features:

• Mel spectrograms: frequency-time representation
• Chroma features: pitch class energy over time
• MFCC: timbral characteristics
• Beat detection and tempo estimation

---

Section C: Songwriting Craft

Songwriting craft guidance and AI music prompts.

Key principles:

• Structure matters: verse-chorus-verse-chorus-bridge-chorus is standard
• Lyrics should match the rhythm and mood of the intended style
• Hook/chorus should be memorable and repeatable
• Bridge provides contrast before final chorus
• For AI generation, be explicit about style, mood, instrumentation, and structure in prompts

---

Pitfalls

1. AI music generation is stochastic — multiple generations may be needed for quality results
2. Lyrics structure affects output — unstructured text produces unstructured music
3. Style tags are powerful — be specific about genre, mood, era, instrumentation
4. Audio analysis requires the actual audio file — cannot analyze from text description alone