42951d11-4a79-47c0-8576-03f8950d62d7

**POSITIVE PROMPT** Plate for shot S1.23 — @Loft, derived from the **P-05 glass macro plate at minimum focus**. Vertical 9:16, 1440x2560. Extreme macro on **@Monitor1's panel substrate**, closer than any previous frame in the episode: **the subpixel structure is visible**, the RGB stripe elements resolving as discrete cells behind the anti-glare coating. The image is abstract — we are inside the screen, not looking at it. Single plane, no depth, no context, no object of the room. Lens character 85mm macro at minimum focus, 0.18 m, camera at seated eye height 118 cm, square to the screen plane. **No reflection exists at this distance** — she has disappeared from her own show. Surface: the anti-glare coating's granular scatter layer sitting over the RGB subpixel grid, the two textures interfering slightly. Fine dust particles at this scale. Cell boundaries dark and even. Matte, W2 panel. Lighting: **self-luminous only.** No external source, no sodium, no key, no fill, no reflection, no ambient. **Generate the panel dark and unlit** — subpixel structure present and readable, but no illuminated content anywhere. Leave the entire field clean as a comp receiver. No waveform, no green trace, no lit cells, no type, no letters, no numbers, no cursor, no UI elements in the generated image — the trace is animated sample-accurate from the locked VO and composited over this substrate, and the green light crawls across the cells as a comp pass. Static. Locked off. Nothing moves. **NEGATIVE PROMPT** reflection, any reflection, mirrored image, face in reflection, person visible, hands, body, room visible, desk visible, bezel, monitor edge, screen border, second monitor, waveform, audio trace, green trace, lit pixels, illuminated content, spectrum bars, equaliser, oscilloscope, visualiser, glow, bloom, halation, light wrap, neon, lens flare, specular highlight, screen glare, blue-white light, warm light, amber light, sodium light, external light source, ambient light, fill light, text, type, letters, words, numbers, timestamps, cursor, UI elements, interface, window, panel UI, icons, buttons, generated screen content, legible anything, watermark, subtitle, deep focus, sharp background, everything in focus, wider shot, recognisable object, depth, perspective, 3D, bokeh, shallow focus falloff across the whole frame, moiré, aliasing artefacts, scan lines, CRT phosphor, CRT curvature, dot mask, halftone dots, print texture, fabric texture, LED matrix, pristine panel, new monitor, no dust, crushed blacks, high contrast, oversaturated, HDR, sharpening halo, grain, film emulation, vignette, colour grade, stylisation, cinematic look, moody, atmospheric, beautiful, aesthetic, curated, camera movement, push in, drift, handheld, dolly, tilt, angled to screen, off-square, horizontal framing, AI artefacts, morphing edges

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.