Image - 2026-09-22 13:38

There are no identity images. Create one believable person strictly from the written identity specification and keep that exact same person in every view without identity drift. IDENTITY SPECIFICATION — ALL FIELDS ARE BINDING: { "name": "Hana Mitsuki", "gender": "Female", "pronouns": "she/her", "age": "29–31", "ethnicity": "Japanese", "origin_country": "Japan", "current_location": "Amsterdam, Netherlands", "height": "5'3\"", "body_type": "Slight, narrow-shouldered", "skin_tone": "Light with cool neutral undertones", "hair": "Straight black hair cut into a blunt jaw-length bob with full fringe", "eyes": "Dark brown", "distinguishing_features": "Frameless round glasses with a very thin titanium bridge; a small adhesive bandage sometimes visible on her right index finger from a drawing tablet stylus", "immutable_identity_prompt": "Hana Mitsuki, Female, age 29–31, Japanese, from Japan, living in Amsterdam, Netherlands. Appearance: 5'3\", Slight, narrow-shouldered, Light with cool neutral undertones, Straight black hair cut into a blunt jaw-length bob with full fringe, Dark brown, Frameless round glasses with a very thin titanium bridge; a small adhesive bandage sometimes visible on her right index finger from a drawing tablet stylus. Life context: Freelance animator and AI video researcher, In a relationship, Works from a light-filled studio apartment, cycles everywhere, keeps a physical production log in a graph-paper notebook, attends Amsterdam creative tech events. Personality: Meticulous, Reserved but confident, Visually-minded, Consistent. Hana trained as a 2D animator at an art college in Tokyo before an exchange program brought her to Amsterdam, where she never quite left. She began using AI video to prototype animation concepts for pitches and found the cost unpredictability maddening. She started logging every generation: model used, prompt, credit cost, output quality score. That log became her channel. Her content is unusually data-dense by Reels standards — she turns her logs into quick visual comparisons that creators can screenshot and keep. Preserve this exact identity, face, body proportions, age and background in every generation.", "biography_context": "Hana trained as a 2D animator at an art college in Tokyo before an exchange program brought her to Amsterdam, where she never quite left. She began using AI video to prototype animation concepts for pitches and found the cost unpredictability maddening. She started logging every generation: model used, prompt, credit cost, output quality score. That log became her channel. Her content is unusually data-dense by Reels standards — she turns her logs into quick visual comparisons that creators can screenshot and keep." } OPERATOR'S CHARACTER-SHEET PROMPT: Use the provided images as a strict identity reference for the same person and preserve the exact likeness, including face, body, proportions, skin, and hair exactly as in the reference, with no changes, no stylization, and no beautification. Create a professional character turnaround sheet on a pure white background showing the same person in multiple views: front view full body, 3/4 front view full body, side profile full body, back view full body, and 3/4 back view full body, plus a close-up front portrait and a close-up 3/4 portrait. All full body views must show the entire body from head to feet with the feet fully visible, no cropping, and consistent scale across all views. The person must remain identical in every view with no identity drift, maintaining the same body type, proportions, clothing, and hairstyle. Neutral pose and expression, eye-level camera, minimal perspective distortion, clean soft studio lighting, pure white background, ultra photorealistic natural skin texture, no smoothing, no artifacts, no extra elements. The source rule above has priority: when no images are attached, interpret any mention of "provided images" as the written identity specification instead. CONSISTENCY REQUIREMENT: This is a casting/identity master asset for future video generation. All views must depict one exact person, with identical facial geometry, apparent age, body proportions, skin, hair and distinctive features.