Generates and edits high-fidelity visual assets using Gemini 3.1 Flash with support for text-to-image, product mockups, and iterative refinement.
This skill acts as a foundational design layer for Claude Code, leveraging Google's Gemini 3.1 Flash Image Preview model to create professional-grade visual assets directly within your development workflow. It streamlines the entire creative process—from initial text-to-image generation and complex image-to-image editing to multi-turn refinements via chat. By enforcing structured prompt engineering and providing precise control over hex colors, typography, and aspect ratios, it ensures consistent, brand-aligned outputs for social media, marketing materials, and web design.
主な機能
011 GitHub stars
02Product-to-graphic transformation and image editing
03Advanced text rendering within generated images
04Multi-turn refinement for iterative design adjustments
05Structured prompt engineering for precise visual control
06Support for multiple aspect ratios and resolutions up to 4K
ユースケース
01Converting raw product photos into polished social media marketing graphics
02Iteratively refining brand illustrations or posters through natural language commands
03Generating consistent UI assets and web banners with specific brand hex codes
