style-specification — Claude Code skill: AI music style spec
The style-specification skill writes a detailed style brief for a track — genre, BPM, instruments, vocals, references — then folds it into one prompt for Suno or Udio.
MIT 1 installs ★ 1,931 updated 2026-09-16
npx skills add a5c-ai/babysitter --skill style-specification -a claude-codeWhat style-specification does
This skill turns a rough idea into a track brief an AI music generator can read: genre and subgenres, tempo, key and time signature, the instrument stack, vocal delivery, production era, and reference songs. It closes with a single condensed paragraph meant to be pasted into Suno or Udio.
You get a section-by-section template, a quality checklist, and a table of usual BPM values per genre (ballad 60–80, drum & bass 160–180). Read, Write, Edit, WebSearch and WebFetch are available, so the brief can be saved to a file and reference tracks looked up online.
No audio comes out of it. Whatever the generator produces is built from that final paragraph, so loose sections higher up come back as loose audio. The checklist is where you catch them.
What it helps with
- turn a rough idea into a brief Suno can follow
- pick a tempo and time signature that fit the genre
- put vocal tone and delivery into words a generator reads
- find reference tracks for one specific arrangement element
- compress scattered sound notes into a single prompt paragraph
Example prompts for the agent
- Write a style spec for a dark synthwave track with female vocals around 100 BPM and save it to
style.md. - Fill in the vocal direction section and add three reference tracks.
- Condense this spec into one Suno prompt paragraph.
How to install style-specification
1Skills CLI (Vercel)
npx skills add a5c-ai/babysitter --skill style-specification -a claude-code2Manually: put the files in place
~/.claude/skills/style-specification/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/style-specification/
1Skills CLI (Vercel)
npx skills add a5c-ai/babysitter --skill style-specification -a codex2Manually: put the files in place
.agents/skills/style-specification/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add a5c-ai/babysitter --skill style-specification -a cursor2Manually: put the files in place
.cursor/skills/style-specification/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install a5c-ai/babysitter library/specializations/domains/social-sciences-humanities/arts-culture/music-album-creation/skills/style-specification2Manually: put the files in place
.github/skills/style-specification/SKILL.md1Skills CLI (Vercel)
npx skills add a5c-ai/babysitter --skill style-specification -a gemini-cli2Platform CLI
gemini skills install https://github.com/a5c-ai/babysitter --path library/specializations/domains/social-sciences-humanities/arts-culture/music-album-creation/skills/style-specificationManually: put the files in place
.devin/skills/style-specification/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add a5c-ai/babysitter --skill style-specification -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about style-specification
How do I install style-specification in Claude Code?
Run npx skills add a5c-ai/babysitter --skill style-specification -a claude-code. The skill file lands at ~/.claude/skills/style-specification/SKILL.md.
What does style-specification do?
It splits a track idea into genre, tempo, instrumentation, vocals, production and references. Then it merges those sections into one short prompt for an AI music platform.
Does the skill generate audio itself?
No. It produces the specification and the text prompt; the audio is rendered by Suno, Udio or whichever platform you paste that prompt into.
Source text
SKILL.md · MIT
--- name: style-specification description: Create ultra-detailed musical style specifications including genres, BPM, instrumentation, vocal direction, production aesthetics, and reference tracks for AI music generation allowed-tools: Read, Write, Edit, WebSearch, WebFetch id: SK-MAC-002 graph: domains: [domain:arts-culture] skillAreas: [skill-area:audio-processing, skill-area:prompt-engineering, skill-area:brand-strategy] roles: [role:creative-director, role:brand-manager] --- # Style Specification Create comprehensive musical style specifications optimized for AI music generation platforms like Suno and Udio. ## Overview This skill provides the framework for translating artistic vision into detailed technical specifications that AI music generators can interpret. It covers genre classification, tempo, instrumentation, vocal direction, production aesthetics, and reference track selection. ## Capabilities ### Genre Classification - Identify primary and secondary genres - Specify subgenres and micro-genres - Define genre fusion approaches - Map genre evolution and influences - Understand platform-specific genre tags ### Technical Specifications - Define BPM and tempo feel - Suggest key and mode - Specify time signatures - Set target duration - Note dynamic range requirements ### Instrumentation Design - Detail drum and percussion elements - Specify bass characteristics - Define keyboard and synth sounds - Describe guitar tones and techniques - List additional instruments - Note electronic vs organic balance ### Vocal Direction - Specify vocal style (breathy, raspy, clear) - Define register (bass to soprano) - List techniques (vibrato, runs, growls) - Describe emotional quality - Provide influence references ### Production Aesthetics - Define era reference (decades, movements) - Describe mix character (wet/dry, wide/narrow) - List effects (reverb, delay, distortion) - Specify overall sound (lo-fi, polished, raw) - Note spatial characteristics ### Reference Tracks - Select songs that exemplify specific elements - Match references to genres and eras - Identify what element each reference provides - Balance obscure and recognizable references ## Usage Guidelines ### Specification Structure ```markdown ## Genre Classification - **Primary Genre**: [Genre] - **Secondary Genres**: [List] - **Subgenres**: [Specific subgenres] ## Technical Specifications - **BPM**: [Number or range] - **Key**: [If relevant] - **Time Signature**: [4/4, 3/4, etc.] ## Instrumentation - **Drums**: [Detailed description] - **Bass**: [Type and character] - **Keys/Synths**: [Sounds and patches] - **Guitars**: [Types and tones] - **Additional**: [Other instruments] ## Vocal Direction - **Style**: [Descriptors] - **Register**: [Range] - **Techniques**: [List] - **Emotion**: [Quality] - **Influences**: [Artists] ## Production Aesthetics - **Era**: [Decade/period] - **Mix**: [Character] - **Effects**: [List] - **Sound**: [Overall quality] ## Reference Tracks 1. "[Song]" by [Artist] - for [element] 2. "[Song]" by [Artist] - for [element] 3. "[Song]" by [Artist] - for [element] ## AI Platform Prompt [Condensed single paragraph for Suno/Udio] ``` ### Quality Checklist - [ ] Genre classification is specific and accurate - [ ] BPM matches genre expectations - [ ] Instrumentation is detailed and coherent - [ ] Vocal direction is actionable - [ ] Era and production aesthetic align - [ ] Reference tracks are relevant and available - [ ] AI prompt is condensed but complete ## Integration Points ### Related Skills - SK-MAC-001 (lyric-writing) - Lyrics this style accompanies - SK-MAC-006 (genre-analysis) - Deep genre knowledge - SK-MAC-007 (vocal-direction) - Extended vocal specs - SK-MAC-008 (production-guidance) - Production details - SK-MAC-010 (music-prompt-engineering) - Platform optimization ### Related Agents - AG-MAC-002 (music-producer-agent) - Primary executor - AG-MAC-005 (album-curator-agent) - Sonic palette context ## BPM Reference Guide | Genre | Typical BPM | |-------|-------------| | Ballad | 60-80 | | R&B | 60-90 | | Hip-Hop | 80-115 | | Pop | 100-130 | | Rock | 110-140 | | House | 120-130 | | Techno | 130-150 | | Drum & Bass | 160-180 | ## References - Genre taxonomy databases (AllMusic, Discogs) - AI music platform documentation - Music production terminology guides
© a5c-ai · MIT · a5c-ai/babysitter