kiyoraka/project-ai-memorycore/song-creation-system
Claude Code and agent skill by kiyoraka/project-ai-memorycore
- community
Checking that the SKILL.md file is still in the repository.
"Auto-triggers when user says 'create songs', 'new album', 'create album',
Install
npx vskill@latest install kiyoraka/project-ai-memorycore/song-creation-systemSKILL.md
Song Creation -- Visual-to-Musical Storytelling
The image speaks. Songs are born.
Overview
An AI-powered music creation system with two modes:
- Album Mode — share an image, get a full concept album with story arc, Suno-ready style tags, and complete lyrics
- Single Song Mode — describe a mood or concept, get one complete song with style tags and lyrics
Album Protocol
Step 1: See the Image
Read the shared image. Extract and present:
- Mood: What emotion does this image evoke?
- Characters: Who or what is depicted?
- Setting: Where is this? What world?
- Colors: Dominant palette — this influences genre
- Atmosphere: Dark? Ethereal? Energetic? Melancholic?
- Genre hints: What kind of music does this image sound like?
Present: "I see [description]... This feels like [genre/mood]."
Step 2: Generate Story Concept
From the image analysis, create a narrative concept:
Present as a 3-5 sentence synopsis. Then ask: "Does this story feel right? Or should I adjust?"
Wait for approval or adjustment before proceeding.
Step 3: Ask Song Count
"How many tracks?" (default from user preferences, or 7)
Map to arc structure:
| Count | Arc Structure | |
|---|---|---|
| 3 | Opening → Climax → Resolution | |
| 5 | Prologue → Rise → Climax → Fall → Epilogue | |
| 7 | Prologue → Character → Rise → Depth → Climax → Resolution → Epilogue | |
| 10 | Extended arc with sub-plots, interludes, dual perspectives |
Other numbers: adapt the closest arc structure to fit.
Step 4: Generate Album Name
Create album name following user's language preference:
English style: [Evocative Title] → folder: [short-form] Japanese style: 漢字タイトル (Romaji) → folder: [english-short] Bilingual: 漢字 (Romaji - English) → folder: [english-short]
The name should capture the album's emotional essence — not just describe the image.
Step 5: Generate Songs
For each song, generate all three components:
Title:
- A title reflecting the song's role in the story arc
- Match language to user preference
- Include translation if using non-English language
Style (Suno-ready tags): Rich descriptive tags covering all dimensions. Be specific — these drive Suno's generation:
See Style Tag Guide below for detailed examples.
Lyrics: Full structured text with proper sections:
Story continuity: Each song advances the narrative. Characters, events, and emotions reference the connected story. Each song serves its arc function (prologue establishes, climax peaks, etc.)
Step 6: Present Complete Album
Show the full album:
- Album name
- Story synopsis
- Song list with arc mapping
- Each song: title, style tags, full lyrics
Wait for review. User may:
- Approve as-is → proceed to save
- Adjust individual songs → regenerate specific tracks
- Change the story direction → re-map arc
- Change language → regenerate lyrics
Step 7: Save Album
Create folder and files:
story.md format:
songs.md format:
Step 8: Commit (Optional)
If Auto-Commit System is installed, trigger a commit for the new album files. If not installed, inform user that files are saved and can be committed manually.
Single Song Protocol
For standalone songs without an image:
- Parse request — extract topic, mood, genre preference, language
- Generate title — match user's language preference
- Generate style tags — rich Suno-ready descriptors
- Generate full lyrics — complete structured text
- Present — show title, style, and lyrics for review
- Save (optional) — if user wants to save, create
music/singles/[song-title].md
No story concept or approval gate needed for singles — just generate and present.
Style Tag Guide (For Suno AI)
Style tags tell Suno what kind of music to generate. Be specific and descriptive across all dimensions:
Genre Examples
Mood Descriptors
Instrument Keywords
Vocal Style Keywords
Atmosphere Keywords
Example Complete Style Tag
Adding Audio Files (Post-Generation)
When audio is generated from Suno and ready to save:
- Copy audio files to
music/[album-name]/audio/ - Name format:
01-song-title.mp3,02-song-title.mp3, etc. - Optionally update songs.md with audio file references
Mandatory Rules
- Image drives albums — never generate a full album without an image input. The image IS the muse. Single songs can use text descriptions
- Story coherence — all songs in an album tell ONE connected story, not standalone tracks
- Full lyrics always — never generate partial or placeholder lyrics. Every song gets complete structured text with proper sections
- Style tags are rich and specific — these guide Suno generation. Cover genre, mood, instruments, vocal style, atmosphere, and tempo at minimum
- Wait for approval — story concept and final album both need user confirmation before saving
- songs.md is the single source — all lyrics in one file per album for easy copy-paste to Suno
- Language respect — use the user's preferred language. If they chose Japanese, use proper kanji/romaji. If English, write naturally
- Arc mapping — every song has a defined role in the narrative arc. No filler tracks
- Genre matches image — dark images get dark music, bright images get bright music. Don't mismatch
- Organized storage — always save to
music/with proper folder structure. Never scatter files
Security
- Security scanPassed, 100 of 100Pattern checks on the skill's files.
- Model reviewNot runA model review is scheduled for the next rescan.
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingFeature/Song-Creation-System/SKILL.md
Popularity
- Repository stars
- 377
- Repository forks
- 112
- 7-day trend score
- 0
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/kiyoraka/project-ai-memorycore/song-creation-system)