What problem does it solve? Writing effective prompts for AI image generators is difficult, especially when trying to recreate an existing image's style, composition, and subject. This Skill reverse-engineers any uploaded image into precise, model-ready prompts so users can reproduce similar results across different AI drawing platforms. ## Core Features & Use Cases - Five-Layer Image Analysis: Systematically extracts subject, environment, lighting, composition, and quality features using a structured checklist. - Multi-Model Output: Converts one analysis into formats for SD1.5, SDXL, Flux, Midjourney, Pony Diffusion, GPT-Image2, and 12+ other models, including positive and negative prompts. - Character & Scene Recognition: Identifies anime characters, celebrities, and 20+ scene types (portrait, landscape, product photography, ink painting, pixel art) with confidence labeling. - Use Case: Upload a cyberpunk cityscape photo and receive ready-to-use prompts for Midjourney (with --ar and --s parameters), Stable Diffusion (with weight syntax and negative prompts), and Flux (natural language format) in one response. ## Quick Start Upload an image and ask the agent to reverse-engineer it into prompts for your target AI drawing models.
