932c0cda-dda0-4108-ad87-fd6de8b91d4d

Create a image [SHOT TL_EP01_SH01.03 | REFS: @Wren → wren_genesis_frontal.png, wren_genesis_threequarter_left.png, wren_genesis_monitorkey.png | @Loft → P-01_frontal_plate.png] @Wren at her desk in @Loft at 02:00, working. Extreme close-up: the frame holds her brow down to her upper lip and nothing else. She is transcribing. Head square to the monitor, not to the lens, face turned to camera-left. Forward head, rounded shoulders, weight settled in the chair. Eyes tracking text left to right in short reading jumps, lids relaxed, gaze locked on the screen plane just off camera-left. Mouth closed and at rest. No reaction of any kind to what she is hearing. Hands out of frame, typing. State: fatigue 30 percent, hair wash-cycle day 3, story clock 02:00. T-zone sheen at the nose and forehead, under-eye at its day-three depth, lower lip dry. No makeup, no polish, no jewellery in frame at this crop. Wardrobe @Wren/EP1-Nightdesk, below the frame line. @Headphones on, the band pressing a flattened channel into the hair above each ear. Hair tucked behind the right ear and loose on the left, unwashed at day three, moving in separated strands, baby hairs standing at the temples, never smoothed. Framing, vertical 9:16: she sits left of centre, facing camera-left, her eyes resting on the upper third line. The top of frame crops the forehead — no headroom. The right third of the frame is dark falloff with nothing readable in it. Single plane, no depth, no desk, no monitor, no room, no background detail. Camera: ARRI Alexa 35, Zeiss Supreme Prime 50mm at T2.0, seated eye level 118 cm, 0.6 m from her, frontal, locked off. Honest perspective, no compression stylisation. Focus on the near eye, background unresolved. Light is inherited from the @Loft P-01 plate and nothing is added to it: one soft source from front and slightly below at 6500K falling off fast, and a weak 2100K rim barely catching the right shoulder and the loose hair at the extreme frame edge. Everything above the brow goes to dark. Blacks lifted and noisy. No fill, no shaping rim, no practical, no haze, no light that is not those two. Texture: unretouched skin with visible pore structure, real oil and dryness, fine lines present at the outer eye, sclera not white, tear film reading as one small low specular at the lower lid. Hair in separated strands, never a sheet. Photorealistic still frame, prestige television production quality, documentary-grade realism, 9:16, 1440 x 2560. Delivered ungraded: no grain, no halation, no film emulation, no stylisation of any kind. NEGATIVE PROMPT beauty-filter skin, retouched skin, smoothed skin, poreless, plastic skin, waxy face, airbrushed, symmetrical face, corrected asymmetry, makeup, mascara, lip gloss, groomed brows, styled hair, volumised hair, hair moving as one sheet, wind, earrings, jewellery, clean skin, young skin, glamour portrait, model, headshot, stock-photo posing, posed to camera, eye contact with lens, looking up, smiling, emotive expression, tears, glassy eyes, dead eyes, crossed eyes, pupils not matching the light, green cast, green key light, coloured light on the face, rim light for shape, fill light, practical lamp, overhead light, ceiling light on, third light source, haze, volumetric light, god rays, lens flare, bloom, glow, halation, crushed blacks, HDR, oversaturated, teal and orange, graded look, film grain, film emulation, vignette, background detail, visible room, visible desk, visible monitor, screen content, text, letters, numbers, interface, watermark, logo, caption, subtitle, headroom, centred subject, symmetrical composition, balanced frame, wide shot, second person, duplicate person, deformed anatomy, warped ear, floating headphones, headphone band not contacting the hair, missing headphone impression, hair tucked on the left, hair loose on the right, cartoon, illustration, 3D render, CGI, video game, oversharpened, low resolution

Generate a single, hyper-realistic character reference image (strictly no collages, text, watermarks, or borders). The image must feature a cinematic, three-quarter portrait of the described character, maintaining a sharp, tack-focused emphasis on the face with natural skin textures, pore details, and realistic micro-expressions. The lighting should be cinematic and directional (e.g., subtle split or Rembrandt lighting) to give depth, blending seamlessly with the environment specified in the character description. The character must have a natural, living presence, looking slightly off-camera or directly at the lens with a subtle, context-aware expression. Ensure the full head, neck, and upper shoulders are completely unobstructed. This image serves as a high-fidelity identity blueprint for temporal-consistent video generation.