How does the Imagine skill improve image quality?

It uses a structured framework that incorporates technical photography specifications and natural language descriptions optimized specifically for Imagen's T5-XXL language model architecture.

Does this skill support negative prompting and aspect ratios?

Yes, it includes built-in patterns for descriptive negative prompts and provides standard aspect ratio configurations like 16:9, 9:16, and 4:3 for different use cases.

Can I add my own custom art styles to the skill?

Yes, the skill is designed for extensibility. You can add new .md reference files to the references directory to define unique visual styles, design principles, and technical descriptors.

What technical camera settings can I control with this skill?

You can specify professional-grade parameters including lens types (macro, wide-angle, portrait), specific lighting setups, camera equipment (ARRI Alexa, 35mm film), and atmospheric effects.

Imagine: AI Image Prompt Architect

Name: Imagine: AI Image Prompt Architect
Author: rfxlamia

byrfxlamia

•

デザインツール

Generates professional, high-fidelity prompts for Google Imagen 3/4 using technical photography specifications and extensible art style frameworks.

Imagine is a specialized Claude Code skill designed to bridge the gap between creative concepts and high-quality AI-generated imagery. It utilizes a structured Subject-Context-Style framework to craft detailed natural language prompts optimized for Google's Imagen models. By incorporating technical photography parameters like focal lengths and lighting conditions, as well as an extensible library of art style references, this skill enables developers and designers to generate consistent, professional-grade visual assets directly within their workflow, ensuring every generated image meets specific aesthetic and technical requirements.

主な機能

0142 GitHub stars

02Technical photography specifications for lenses, lighting, and equipment

03Built-in support for aspect ratios, safety controls, and text-in-image constraints

04Extensible art style library with dynamic reference-based generation

05Structured Subject-Context-Style prompting framework

06Natural language optimization for Imagen 3/4 T5-XXL models

ユースケース

01Creating high-fidelity stock photos and product mockups with precise camera and lighting settings.

02Rapidly iterating on visual concepts for marketing materials or UI/UX placeholders directly from the CLI.

03Generating consistent character designs and environment art using specific artistic reference files.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add rfxlamia/claude-skillkit imagine

For use in Claude.ai and ChatGPT

Download Skill