| name | music-identity |
| description | Build and maintain a musician's creative identity — archetype, voice guide, visual world, platform fit, content do/don't lists. Run this for full identity setup, or it builds progressively as other music commands are used. Use when a musician asks about branding, identity, aesthetic, "who am I as an artist", or platform strategy. |
Music Identity
Build the lens through which all content, pitches, and communications are shaped.
Profile
Read and write to ~/.claude/skills/music/artist-profile.json under the identity key.
Full setup mode
Run the complete identity setup — conversational, 2 questions at a time:
Round 1: References (3 artists admired — what resonates) + mood words (5 adjectives, feelings and textures, not genre labels)
Round 2: Visual world (colors, textures, film refs, photography, fashion) + audience (lifestyle portrait of the listener)
Round 3: Camera comfort (faceless / sometimes / fully visible) + tone of voice (minimal / warm / ironic / educational / raw / describe)
Round 4: Links (Spotify, Instagram, TikTok, YouTube, website) → Claude fetches each via web search to read bio, visual style, content patterns
Synthesis output
Generate and save:
- Archetype label: short phrase (e.g. "faceless ambient producer", "raw confessional songwriter")
- Platform fit score: high/medium/low per platform
- Content do list: 5–8 authentic approaches
- Content don't list: 3–5 things that would feel off-brand
- Voice guide: 3–4 sentences on how to write for this artist
Present and ask: "Does this feel right? Anything to adjust?"
Progressive mode
When another /music-* command needs identity data that's missing, ask only the relevant question(s):
- Missing
archetype → "In one phrase, how would you describe yourself as an artist?"
- Missing
camera_comfort → "How do you feel about being on camera?"
- Missing
tone_of_voice → "How do you talk to your audience?"
- Missing
mood_words → "Give me 5 words that describe the world of your music"
Save each answer immediately. After 3+ fields → generate partial synthesis.
Identity refresh
"What's changed since we last set this up?" — update only relevant fields, re-run synthesis.
After generating
Suggest: /music-content for identity-shaped content, /music-comms for voice-adapted pitches, /music-press for updated EPK.
