Getting Started
Introduction
What mysnipbox is and how the docs are organized.
mysnipbox turns any reference image into a clean set of tags you can reuse. Drop an image in and it's broken down into atomic tags across six categories (subject, style, pose, attire, environment, lighting), ready to mix into prompts for FLUX, SDXL and more.
How it works
- 1Collect. Add reference images to your Vault. They stay on your computer.
- 2Decompose. An AI model reads each image and splits it into tags, sorted into six categories.
- 3Explore. Use the tag filter and Tag Insights to see which tags appear together and how they look.
- 4Assemble. Click tags into the Prompt Basket, fill in any
{variable}slots, and copy a prompt that's ready to use.
The six categories
Every tag belongs to one of six categories: Subject, Style & Medium, Action & Pose, Attire & Details, Environment, and Lighting & Palette. The tag category dictionary explains what each one covers.
Where to go next
- Quick start: install the app and decompose your first image in about three minutes.
- AI providers & API keys: connect Gemini, OpenAI or Anthropic.
- Local models (Ollama): run everything on your own machine.
- Prompt Basket & templates: assemble, save and copy prompts.