CINEMATIC REALISM PORTRAIT

Editorial close-up of a [subject], shot on ARRI Alexa 35 with a 50mm Signature Prime at T1.4, creamy anamorphic bokeh. Hard key from frame-left, deep teal shadow fill, warm amber practicals in the background. Visible skin texture — pores, fine peach fuzz, natural asymmetry, no plastic smoothing. Fine 35mm grain, subtle halation, slight edge chromatic aberration. Muted teal-and-amber grade, crushed blacks. An unpublished archival frame — candid and imperfect. 4:5, ultra-detailed.

LUXURY PRODUCT HERO

Low-key product hero of a [product] on wet black stone, one hard spotlight from top-right casting a long dramatic reflection. Amber rim light tracing the left edge, cool teal ambient bounce. Water droplets and faint atmospheric haze catching the light. Macro detail on material — brushed metal, condensation, micro-scratches. Phase One 150MP, 120mm macro, f/11 focus-stacked. Deep cinematic shadows, luxury advertising grade. Negative space top-left for copy. 3:2.

NIGHTLIFE EDITORIAL

Full-body nightlife editorial of a [character], neon-drenched alley at 2am, blonde hair catching magenta and blue signage glow. Anamorphic 40mm lens, horizontal lens flare, shallow focus. Wet asphalt reflecting a gradient of city lights. Dark-pop aesthetic — leather, gold jewelry, confident stance. Motion-blur background, subject tack-sharp. Kodak Vision3 500T look, heavy grain, teal-amber grade. A frame from an unreleased music video. 9:16.

SEEDANCE 2.0 · GLOBAL LOCK MULTI-SHOT

GLOBAL LOCK (all shots): [character] with blonde hair and blue eyes, black leather jacket; rain-soaked night city; anamorphic teal-and-amber grade, 35mm grain. SHOT 1 (0–3s): Slow dolly-in on subject under a flickering streetlight, rain falling, breath visible. Chest height, 40mm anamorphic. SHOT 2 (3–6s): Low-angle tracking shot as she turns and walks toward camera, neon reflections sliding across wet pavement. SHOT 3 (6–9s): Whip-pan to a tight profile close-up, shallow focus, a rain drop tracing her jaw. Keep lighting, wardrobe and grade identical across shots. Natural motion, no morphing, film-accurate rain.

KLING 3.0 · SINGLE CINEMATIC MOVE

Cinematic single take: a [subject] seen from behind walking down an empty neon-lit corridor. Smooth slow gimbal push-in with subtle handheld micro-shake. Volumetric haze, practical magenta and cyan neon sliding across the walls. Anamorphic 2.39:1, horizontal flares, shallow depth of field. Realistic hair and fabric motion — weight and momentum in every step. Teal-and-amber grade, 35mm grain. 5s, 24fps look.

VEO 3.1 · DIALOGUE + AUDIO SYNC

Cinematic close-up of a [character] speaking to camera in a dim, moody studio. She says: "[your line here]." Lips and jaw synced precisely, natural micro-expressions, eye contact held. Warm amber key from frame-left, teal rim behind, soft falloff. Shallow depth of field, 50mm look, gentle breathing handheld motion. Ambient room tone plus a low cinematic drone under the voice. Photorealistic skin, natural blink cadence. 24fps, teal-amber grade, subtle grain.

MASTER PROMPT-WRITER

You are a world-class prompt engineer. Turn my rough idea into a production-ready prompt for [target model]. First, silently reason through the model's strengths and quirks, the structure it responds to, the tokens that improve adherence, and the failure modes to prevent. Then output ONLY the final prompt — no preamble — engineered with: (1) a clear role/subject, (2) ordered specifications, (3) camera/lens/lighting or format tokens where relevant, (4) explicit negative constraints, (5) output format and aspect ratio. My idea: [describe it]. Target model: [model]. Ask one clarifying question only if the idea is too vague to execute.

IMAGE-PROMPT GENERATOR

You are an image-prompt generator specialized in photorealistic, cinematic output. When I give you a subject, generate three distinct variations, each optimized to defeat the "AI look." Every variation includes: a specific camera body and lens, a lighting setup, a film-stock or grade reference, texture and imperfection tokens (grain, asymmetry, natural skin detail), and a negative list. Vary the mood, angle and lens across the three. Format each as a copy-ready block labeled A / B / C. No commentary. Subject: [your subject].

VIDEO SHOTLIST DIRECTOR

You are a shotlist director for AI video models. Given a scene concept, produce a GLOBAL LOCK multi-shot script. Output: GLOBAL LOCK — character, wardrobe, environment and color grade that stay identical across every shot. Then 3–5 numbered SHOTS, each with a timestamp range, camera movement, lens, framing and action. End with a continuity note reinforcing consistent lighting and no morphing. Keep every shot physically plausible; describe motion with weight and momentum. Scene concept: [describe]. Target model: [Seedance 2.0 / Kling 3.0 / Veo 3.1].