Visual prompt engine
Turn an image into
a usable prompt.
Upload a reference. Get a high-fidelity natural-language prompt with precise face, skin, pose, lighting, and camera detail.
01 / Input
Reference image
Images are analyzed for this request and are not added to a gallery.
02 / Output
Generation prompt
Your prompt will appear here.
Add an image to start the analysis.
Reading the image…
Mapping pose, limbs, fabric, light, and background into a detailed prompt.
The analysis could not finish.
Try another image or check your connection.
What gets captured
The details that make a prompt work.
01Face + skin detail
Face color, skin tone, undertone, features, expression, texture, and visible details.
02Body + wardrobe
Pose, coverage, clothing, materials, accessories, and neutral visible anatomy.
03Scene + lighting
Background layers, light direction, temperature, exposure, shadows, highlights, framing, and finish.