DIRECTIVE — 30s cinematic single continuous take, 16:9, 24fps filmic motion cadence.
SUBJECT & STORY — [FORMAT] 16:9 aspect, 24fps, ARRI Alexa 35 capture, 17 stops of latitude, creamy highlight rolloff, natural skin rendition, Cooke S4/i glass, warm 'Cooke look', gentle skin falloff, soft creamy bokeh, 40mm near-normal, natural human perspective, mild depth separation, T2.8 aperture, photoreal live-action cinematography, real human skin texture, natural imperfections
[SCENE] EXT./INT. YOUTH HUB — MORNING. LOCATION LOCK — YOUTH HUB — INTERIOR AT NIGHT: The Youth Hub inside at night, warm interior light, and the same building from outside at night with every floor lit — the exact reverse of the opening aerial. Photorealistic observational documentary cinematography, Kurdistan Region of Iraq, 2020s. Real Kurdish faces from Duhok, Sulaymaniyah, Erbil and Halabja — never generic international casting, never Gulf, Levantine or Turkish features. No makeup on anyone, no stylist hair, no whitened teeth, natural skin texture with pores and flyaway hair. Muted grade: warm skin, cool blue shadows, dust. Anamorphic 2x, shallow depth, handheld only where noted. No text, no captions, no logos, no watermark. This exact set, unchanged, in every shot of the scene. The film states plainly, for the first time, what happened: they came back.
[SUBJECTS] HEVI — HEVI — 19-year-old Kurdish woman from Duhok. 1.62 m, slight build, narrow shoulders. Oval face with childhood still in the cheeks, light olive skin. Thick natural untouched eyebrows. Wide dark-brown eyes, long lashes. Dark-brown wavy hair falling just past the shoulders. ZERO makeup in every shot. CONSISTENCY MARK: one small mole about one centimetre below her LEFT eye. Moves too quietly, so nobody wakes. Touches the lanyard at her neck without noticing. Photorealistic observational documentary cinematography, Kurdistan Region of Iraq, 2020s. Real Kurdish faces from Duhok, Sulaymaniyah, Erbil and Halabja — never generic international casting, never Gulf, Levantine or Turkish features. No makeup on anyone, no stylist hair, no whitened teeth, natural skin texture with pores and flyaway hair. Muted grade: warm skin, cool blue shadows, dust. Anamorphic 2x, shallow depth, handheld only where noted. No text, no captions, no logos, no watermark. — wearing Dark jeans, a plain simple white shirt, hair now tied back properly, a Volunteer.KRD lanyard around her neck. No makeup, no jewellery. (identity locked to reference, do not alter face, hair, age or build)
[BLOCKING] ONE CONTINUOUS TAKE: this 30-second clip is a single uninterrupted shot. No cuts, no edits, no dissolves, no jumps in time or place — the camera never stops rolling. The camera may move — dolly, pan, reframe, follow the actors — but it stays one operator, one roll, one piece of film. Play the action straight through in real time so the take can be cut afterwards in the edit. OPENING SCENE: this take establishes the film's world, light and grade — every later scene will match it. It has to hand over cleanly to the next scene — The job. Said out loud, so nobody has to guess. — so end the take on a frame that can cut straight into it. [REFERENCE MAP — bind each reference image to exactly one thing and never mix them] @image1 = Hevi (lead) — canonical face, hair, body and age of this actor; @image2 = Hevi's ONLY approved wardrobe — Hevi — signature day outfit; exact garments, materials, accessories, fit and colors; @image3 = the LOCATION of this scene — Youth Hub — interior at night (set, not a person). Each @image is that one subject only: do not merge two references into a new person, do not swap who is who, and do not invent anyone who is not in this map. [CAST + WARDROBE LOCK — SOLE AUTHORITY] The reference map contains the canonical identity portrait and the ONE approved wardrobe plate for every actor: Hevi (lead). Reproduce those exact faces, hairlines, bodies, skin tones, ages, garments, accessories, materials, fit and colors for the whole take and every scene. Never re-cast, re-age, beautify, re-style, change clothes, remove accessories, blend identities or invent a face. LIVED-IN WORLD: this is a real place on a real day, never an empty set. Populate the exterior with background life at cinematic depth: pedestrians crossing and reacting, parked and moving cars, taxis, a police cruiser, delivery vans, street signage, traffic lights, litter, wet asphalt reflections. Background people are unnamed extras: out of focus or turned away, correctly scaled, never mugging at camera and never resembling the named cast. The named cast in the reference sheets stay the foreground subjects, sharp and clearly readable at all times. ABSOLUTE CONTINUITY: same wardrobe, same light, same set, same time of day for the entire take.
[CAMERA] ARRI Alexa 35 capture, 17 stops of latitude, creamy highlight rolloff, natural skin rendition, Cooke S4/i glass, warm 'Cooke look', gentle skin falloff, soft creamy bokeh, 40mm near-normal, natural human perspective, mild depth separation, T2.8 aperture
[LIGHT] motivated night light in a all floors lit space, practical sources visible in frame, soft falloff into shadow
[ACTION] ACTION (one take, straight through):
The film states plainly, for the first time, what happened: they came back.
[DIALOGUE] The characters speak their lines out loud on camera in English, with accurate lip sync, natural breathing and room acoustics. Generate the spoken dialogue as part of the clip audio. No narration, no subtitles, no on-screen text.
[GRADE] quiet, observational, restrained, human, hopeful without sentimentality, film-grade contrast curve, protected highlights, no digital clipping
[NEGATIVE] identity drift, different face between shots, wardrobe change mid-scene, extra fingers, deformed hands, warped faces, duplicated people, subtitles, captions, watermark, logo, text overlay, cartoon, 3d render, video game look, plastic skin, over-smoothed skin, oversaturated colours, fisheye distortion, motion blur smear, low resolution
CINEMATOGRAPHY — shot on ARRI Alexa 35 with anamorphic primes, T2.0, shallow depth of field, organic lens flares, motivated practical lighting: soft key, negative fill, hard rim for separation. Filmic contrast curve, rich blacks with retained shadow detail, subtle 35mm grain, natural skin tones, no CGI plastic sheen.
GRADE — low-key cinematic colour science, teal-leaning shadows, warm practical highlights, film-print halation on speculars.
CAMERA — one unbroken take, micro-handheld breathing, no cuts, no transitions; performance carries the shot.
SOUND — diegetic in-camera ambience, close-mic detail on the hero action, no music bed unless specified, no voice-over unless specified.
AVOID — a single slow drifting take, static talking-head boredom, warped hands or faces, morphing geometry, on-screen text, watermarks, logos, subtitles, stock-footage look, over-saturated HDR.