English | 日本語
Struggling to manage your AI image generation prompts?
- Prompts get long and you lose track of what's where
- "Where did that prompt go that worked so well last time?"
- Want to change just the hair color, but have to copy-paste and edit everything...
- Similar prompts scattered everywhere, impossible to organize
The more you use AI image generation, the longer and more complex your prompts become. Managing them as plain text has its limits.
Before (Natural Language)
young Japanese woman with long straight chestnut hair and brown eyes,
wearing yellow t-shirt and indigo jeans, standing, peace sign,
simple red background, professional lighting
Where does the outfit end? How do you change just the hair color?
After (YAML Structure)
subject: person
demographics:
gender: woman
age: young
ethnicity: Japanese
appearance:
hair:
length: long hair
style: straight hair
color: ${hair_color} # ← Use variables for easy changes
outfit:
top: t-shirt
bottom: jeans- Hierarchical structure keeps things organized and easy to navigate
- Inheritance & composition lets you reuse common parts
- Variables make dynamic value changes a breeze
A desktop app for easily creating and managing "structured prompts".
| Feature | What it does |
|---|---|
| Template Inheritance | Use _base to share common settings across files, only manage the differences |
| Layer Composition | Use _layers to freely combine outfit, lighting, pose modules |
| Variable System | ${varName} lets you switch hair color, background, etc. with one click |
| Autocomplete | Dictionary-based input completion with source context display |
| Dictionary Manager | Add, edit, delete, import/export dictionary entries from within the app |
| Multi-tab & Split | Open multiple files in tabs, edit side-by-side with split view |
| ComfyUI Integration | Select workflow → Generate → Image generation, gallery management |
| LLM Enhancer | Ollama integration to transform YAML into natural language prompts |
- People who want to systematically manage their prompts
- People who want to create variations of the same character
- ComfyUI users who want to streamline prompt creation
Download the latest version from Releases.
| OS | File |
|---|---|
| Windows | .msi or .exe |
| macOS | .dmg |
- Launch the app
- Select a sample template from the left pane
- Review and edit the content in the central editor
- If there are variables, enter values in the bottom-left form
- If ComfyUI is connected, click "Generate" to create images
See the Tutorial for more details.
- Tutorial - Start here if you're new
- Manual - Complete feature reference
- YAPS Specification - Structured schema details
| Area | Technology |
|---|---|
| Desktop | Tauri 2 |
| Frontend | Next.js 16 + React 19 |
| Editor | Monaco Editor |
| Database | SQLite (tauri-plugin-sql) |
| UI | shadcn/ui + Tailwind CSS v4 |
npm install # Install dependencies
npm run tauri:dev # Development mode
npm run tauri:build # Build
npm run lint # Lintsrc/ # Next.js frontend
src-tauri/ # Tauri Rust backend
data/ # Bundled resources (templates, dictionary, snippets)
docs/ # Documentation
MIT