Prompt AI per Halloween

Use the uploaded selfie as the primary reference. Preserve the subject recognizable face, skin tone, age and natural anatomy. Add subtle cobweb eye makeup, a black turtleneck and soft autumn fog. Use a warm key light, a cool rim light and a dark studio background. Keep realistic skin texture and change only makeup, lighting and atmosphere. Do not reshape the face, add people, add gore, add text or add a watermark.

Use the uploaded photo as the primary reference. Preserve the subject recognizable facial identity and natural proportions. Create an original, fully opaque Halloween costume with deep plum fabric, silver embroidery and structured shoulder details. Place the subject in a moonlit conservatory with fog and warm lanterns. Use photorealistic editorial photography and realistic fabric texture. Keep exactly one person. Avoid copyrighted characters, logos, text and watermarks.

Use the uploaded pet photo as the primary reference. Preserve the animal breed, fur markings, eye color and natural anatomy. Add a soft pumpkin costume with a green leaf collar, without covering the eyes. Place the pet in a warm autumn studio with miniature pumpkins. Create a photorealistic pet portrait with natural paws and fur texture. Avoid extra animals, text and watermarks.

Use Image 1 and Image 2 as separate identity references for two adults. Preserve both faces and body proportions. Create coordinated, fully opaque Halloween costumes with equal prominence and believable contact shadows. Place them in a desert dusk scene with warm string lights. Keep exactly two people and do not swap or merge faces. Avoid extra people, distorted hands, logos, text and watermarks.

Create the still portrait first: one fully clothed adult in an original gothic Halloween outfit, moonlit garden, soft fog and a calm expression. Use photorealistic editorial photography with stable anatomy and natural fabric texture. Keep the face clearly visible and centered. No text, logos or watermarks. After generating the still, open the image-to-video workflow to animate a slow head turn and drifting fog.