image reference prompt engineer

Turn any reference image into accurate, detailed, copy-ready AI image prompts and reusable templates. Full Description Image Reference Prompt Engineer turns uploaded reference images into professional, detailed and copy-ready prompts optimized for ChatGPT Image. Instead of manually trying to identify an image's style, composition, lighting, colors, camera angle, textures and visual details, this skill analyzes the reference and converts its visible characteristics into a structured generation prompt. It is designed for creators, designers, prompt engineers and AI image users who want to recreate a visual direction, extract a reusable style or build prompt products from reference images. What this skill can do Analyze uploaded reference images. Reverse-engineer their visual characteristics into AI prompts. Identify subject, composition, perspective, environment and framing. Extract colors, lighting, textures and rendering style. Describe typography and graphic elements when visible. Generate clean, copy-ready prompts optimized for ChatGPT Image. Extract only the visual style while allowing you to replace the original subject. Create creative variations while preserving the visual direction. Convert references into reusable PromptBase-style templates. Automatically create editable [--variable--] sections for prompt products. Analyze multiple reference images and combine their recurring visual characteristics. Add useful negative constraints without bloating the prompt. Multiple output modes The skill automatically adapts its response depending on what you need: Final Prompt Mode — Returns only the finished copy-paste prompt. Visual Analysis Mode — Explains the composition, style, palette, lighting and important visual characteristics before generating the prompt. PromptBase Template Mode — Converts the reference into a reusable prompt template containing 2–5 editable variables. Style Extraction Mode — Extracts the visual aesthetic without copying the original subject. Creative Variation Mode — Keeps the visual language of the reference while changing selected elements such as the character, environment, colors or action. Built for accurate reference analysis The skill focuses on details that are actually visible in the reference image and avoids inventing unsupported elements. It analyzes: Main subject and physical characteristics Clothing, accessories and props Pose and expression Image composition Camera angle and perspective Foreground and background Environment Visual style Linework and rendering Color palette Lighting direction Shadows, highlights and reflections Textures and materials Typography and graphic elements Aspect ratio and image format Why use this skill? Creating a strong AI image prompt from a reference can require a surprising amount of visual analysis and prompt engineering. This skill turns that process into a repeatable workflow. Upload a reference image, explain what you want to preserve or change, and receive a structured prompt ready to use. It is particularly useful for: AI artists Prompt sellers PromptBase creators Graphic designers Concept artists Social media creators Character designers Product illustration creators Users building consistent AI-generated visual styles The skill produces text prompts by default and does not automatically generate images, giving you complete control over how and where the resulting prompts are used. Optimized for ChatGPT with image input and ChatGPT Image.