prompt-1
Words spoken and shown on screen: # 🎬 أقنعة الحديد - MASKS OF IRON Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-2
text
prompt-2
## الفيلم القصير · الجزء الأول (0:00 - 5:00) **السلسلة:** Coach ProMedic - عرض سينمائي بلا وجوه · **المدة:** 5 دقائق · **البنية:** 5 أحداث × 4 مشاهد × 15 ثانية = 20 مشهد **المصادر المعاد توليدها وتطويرها:** محتوى تطبيق king.addict.best (Coach episodes + Coach content: MASKS OF IRON, OBSESSIVE/SCROLL/SCALE, Iron Log) + ملفات coach-ready-scripts (16-9, part-1, part2) · محسّنة بمهارات: cinematic-prompt-guide + Cinematic prompts v3 + Attractive elements ultimate + Media-production skill --- Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. Promedic1 picture rule: draw no cross and no snake anywhere in the image. No medical cross, no plus-sign cross, no hospital cross, no snake, no serpent, no Rod of Asclepius. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-3
text
prompt-3
## ⚙️ قواعد الإنتاج الثابتة (تقرأ مرة واحدة - تسري على كل مشهد) **STYLE-LOCK (تُلصق في نهاية كل برومبت صورة):** ``` realistic cinematic film still, photorealistic live-action rendering, natural dramatic lighting, subtle film grain, 35mm anamorphic lens, shallow depth of field, 16:9, ultra-detailed 4K. FACELESS RULE: every character is a realistic human with a completely blank smooth featureless face - NO mask object, NO oval, NO circles on the face - the Arabic nickname text glows directly on the blank face surface. NOT manga, NOT anime, NOT illustration ``` **GRADE-PHASE لهذا الجزء (الفصل الأول = القاع):** ``` desaturated steel-blue world (#4A5568), concrete grays, harsh fluorescent whites; the ONLY vivid colors: face-text colors + electric cyan (#00D9FF) app glow ``` **NEGATIVE-CHECKLIST (يُرفق مع كل توليد - تأكد من غيابها):** ``` deformed hands, extra fingers, fused fingers, missing fingers, mutated limbs, extra limbs, disproportionate anatomy, bad anatomy, asymmetrical body, visible eyes, visible nose, visible mouth, visible eyebrows, realistic facial features, eye-socket shadow, nose shadow, lip shadow, blank face with eyeholes, oval outline, circles on the face, transparent visor, blurry, low resolution, pixelated, jpeg artifacts, oversaturated noise, watermark, signature, random text, garbled text, duplicate character, cloned figure, extra character, cropped, out of frame, low detail background ``` **قواعد صارمة:** - وجوه = أسطح صمّاء تمامًا: لا عيون، لا أنف، لا فم، لا حواجب - فقط نص اللقب العربي على سطح الوجه الصامد. الانفعال يُروى بلغة الجسد + خطوط المانجا (سرعة، عرق على سطح الوجه الصامد، اهتزاز، هالة). - **ممنوع الموسيقى وكل صوت آلي موسيقي - أصوات طبيعية فقط** (مطر، أزيز نيون، ارتطام حديد، تنفّس، خطوات، نقرات تطبيق). التطبيق "يصدر" أصوات UI رقمية طبيعية ضمن عالم القصة. - كل شخصية = لقب عربي واحد قصير مضيء على سطح وجهها الصامد. نص اللقب على سطح الوجه هو "وجه" الشخصية: يهتز عند الخوف، يتوهج عند الانتصار، يتشقق تحت الضغط. - **سير العمل:** ولّد SECTION ZERO أولًا → احفظ كل بطاقة كصورة مرجعية `REF-<name>.png` → في كل مشهد استخدم الصورة المرجعية + الوصف الموجز. - نص عربي على الوجوه الصمّاء: إن فشل التوليد الحروفي العربي، استخدم البديل اللاتيني المذكور في كل بطاقة (نفس الحجم والموضع). --- Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-4
text
prompt-4
Words spoken and shown on screen: # 🔲 SECTION ZERO - بطاقات توليد الشخصيات > ولّد البطاقات الخمس أولًا، كل بطاقة على حدة، 3-4 نسخ واختر الأنظف نصًّا. هذه الصور هي المراجع الثابتة لكل المشاهد في الأجزاء الثلاثة. Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-5
text
prompt-5
### البطاقة 1 - منحني (KENJI - THE SLOUCHER) **بوكس التوليد:** ``` Full-body character reference sheet, single character centered, full frontal view. A tall young man, severely hunched shoulders, neck craned down, hands pressing his lower back. Completely featureless smooth off-white ceramic-smooth blank face surface, hugging the natural head contour - NO eyes, NO nose, NO mouth, NO eyebrows, absolutely blank surface - with the Arabic word «منحني» written in a drooping curved blue font (#4299E1) across the upper blank face, the blank face has no other features, only the text «منحني» (Latin fallback: "BENT" in the same drooping blue font). Messy black hair falling over face edges. Oversized gray hoodie with hood up, baggy dark sweatpants, medical posture brace visible under the hoodie. Realistic cinematic film style, photorealistic, dramatic natural film lighting, subtle film grain, neutral gray studio background, full-body sharp focus, 16:9. NO eyes, NO eye holes, NO facial features, NO mask eyeholes, NO oval outline. ``` **الوصف الموجز (يُذكر في كل مشهد):** `REF: منحني - طويل محدوب، وجه أبيض صامد بعنوان أزرق منحنٍ، هودي رمادي، يضغط أسفل ظهره` **الشخصية:** شاب عشريني، عموده الفقري ينذر بالانهيار، كبرياؤه أكبر من ظهره. بطله الكوميدي-المأساوي: يرفض الضعف فيتكسّر. Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-6
text
prompt-6
### البطاقة 2 - باهتة (YUKI - THE FADED) **بوكس التوليد:** ``` Full-body character reference sheet, single character centered, full frontal view. A thin frail woman wrapped in an oversized beige cardigan three sizes too big, fingerless gloves, a loose scarf. Completely featureless smooth pale-gray matte blank face surface, hugging the natural head contour - NO eyes, NO nose, NO mouth, NO eyebrows, blank surface - with the Arabic word «باهتة» in a thin washed-out lavender font (#B794F4), slightly smudged, the blank face has no other features, only the text «باهتة» (Latin fallback: "FADED" thin lavender font). Thin brittle shoulder-length dull brown hair. Slouched fragile posture, one hand steadying herself on a wall. Realistic cinematic film style, photorealistic, dramatic natural film lighting, subtle film grain, neutral gray studio background, full-body sharp focus, 16:9. NO eyes, NO eye holes, NO facial features. ``` **الوصف الموجز:** `REF: باهتة - نحيلة هشّة، وجه رمادي صامد شاحب بعنوان لافندر باهت، كارديجان بيج ضخم ووشاح` **الشخصية:** طاقتها وذاكرتها وشعرها تتساقط. تظن أن "الأكثر" هو الحل - درسها سيكون "الجرعة الصحيحة". Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-7
text
prompt-7
### البطاقة 3 - ضخم (TAKESHI - THE HEAVY) **بوكس التوليد:** ``` Full-body character reference sheet, single character centered, full frontal view. A massive obese man 140kg+, heaving chest, sweat stains on a tight white t-shirt, stretched black gym shorts, knee wraps, a giant water bottle in one hand. Completely featureless smooth dark-red glossy smooth blank face surface, hugging the natural head contour - NO eyes, NO nose, NO mouth, NO eyebrows, blank surface - with the Arabic word «ضخم» in bold compressed yellow font (#ECC94B) stretched vertically, the blank face has no other features, only the text «ضخم» (Latin fallback: "HEAVY" bold yellow). Short sweaty hair, receding hairline. Wide heavy stance, wheezing posture. Realistic cinematic film style, photorealistic, dramatic natural film lighting, subtle film grain, neutral gray studio background, full-body sharp focus, 16:9. NO eyes, NO eye holes, NO facial features. ``` **الوصف الموجز:** `REF: ضخم - عملاق متعب، وجه أحمر صامد داكن لامع بعنوان أصفر مضغوط، تي شيرت أبيض ضيق وزجاجة مياه` **الشخصية:** كان يجري 10 كيلومترات. دفن نفسه تحت 60 كيلو من الندم. مرآته تحمله سرًّا سيفضحه الفيلم. Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-8
text
prompt-8
### البطاقة 4 - شرارة (HANA - THE SHRUNKEN) **بوكس التوليد:** ``` Full-body character reference sheet, single character centered, full frontal view. A tiny fierce woman 148cm tall, child-proportioned limbs, disproportionately large platform training shoes, pink crop top that swallows her frame, compression leggings. Completely featureless smooth peach-colored smooth blank face surface, hugging the natural head contour - NO eyes, NO nose, NO mouth, NO eyebrows, blank surface - with the Arabic word «شرارة» in a tiny but fierce red font (#E53E3E) stacked in two rows, the blank face has no other features, only the text «شرارة» (Latin fallback: "SPARK" fierce red). Short fiery orange-red pixie hair. Fists clenched, wide confident fighting stance. Realistic cinematic film style, photorealistic, dramatic natural film lighting, subtle film grain, neutral gray studio background, full-body sharp focus, 16:9. NO eyes, NO eye holes, NO facial features. ``` **الوصف الموجز:** `REF: شرارة - قصيرة شرسة، وجه خوخي صامد بعنوان أحمر ناري من سطرين، شعر برتقالي ناري، قبضتان مشدودتان` **الشخصية:** لا يصلح لها جهاز ولا مرآة ولا مقعد - فتقف على صناديق. هي قلب الكوميديا في الفيلم وشرارة الأمل. Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-9
text
prompt-9
### البطاقة 5 - المعلم (REI - THE COACH) **بوكس التوليد:** ``` Full-body character reference sheet, single character centered, full frontal view. A 45-year-old muscular man with an atrophied left shoulder, wearing a vintage tracksuit with a faded logo, holding an empty clipboard and a broken stopwatch. Completely featureless smooth weathered black-leather smooth blank face surface, hugging the natural head contour - NO eyes, NO nose, NO mouth, NO eyebrows, blank surface - with the Arabic word «المعلم» in cracked white font like old gym chalk, one edge of the old text layer visibly peeling, the blank face has no other features, only the text «المعلم» (Latin fallback: "MENTOR" cracked chalk-white). Silver-gray hair slicked back with a widow's peak. Tired upright posture, one hand rubbing his temple. Realistic cinematic film style, photorealistic, dramatic natural film lighting, subtle film grain, neutral gray studio background, full-body sharp focus, 16:9. NO eyes, NO eye holes, NO facial features. ``` **الوصف الموجز:** `REF: المعلم - ربّان متعب بكتف ضامر، وجه بلون الجلد الداكن الصامد أسود بعنوان طبشوري متشقق وزاوية مقشّرة، تدريب قديم وحافظة فارغة` **الشخصية:** درّب أبطالًا ثم خسرهم كلهم بالورق والفوضى. قناعه يخفي تحت زاويته المقشّرة لقبًا قديمًا - «الفاشل». الفيلم سيكشفه في الجزء الثالث. --- Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-10
text
prompt-10
Words spoken and shown on screen: ## الحدث ١ - «تحت الصفر» (0:00 - 1:00) > خطاف البداية: عمود فقري ينهار على شاشة أشعة. أربعة أشخاص عند القاع، ومدرّب خسر كل شيء. كوميديا سوداء خفيفة عبر ضخم وشرارة. Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-11
text
prompt-11
### المشهد 1 (0:00-0:15) - «التشخيص» **BOX A - IMAGE GENERATION:** ``` [REF: منحني - hunched tall figure, off-white blank face «منحني» drooping blue text, gray hoodie] Medium shot from behind, KENJI seated on a sterile medical examination table, hood up, hands on knees, back to camera. On the wall ahead: a glowing X-ray display of a severely curved spine with red alarm circles around lumbar vertebrae. Doctor's dark silhouette at frame edge pointing at the X-ray. Cold blue-white fluorescent light, buzzing tube overhead. + STYLE-LOCK + desaturated steel-blue medical palette. ``` **BOX B - ANIMATION / PRODUCTION / EDIT:** ``` 15s scene. Slow push-in over 10s toward the X-ray; red alarm circles pulse twice like a heartbeat; the fluorescent tube flickers once at second 12 (subtle frame darken). Film grain shimmer. Sound (natural only): fluorescent buzz, paper rustle, one muffled heartbeat-like X-ray pulse tone. End on held frame: red circles frozen. Cut by hard transition to white. ``` **التعليق الصوتي:** - **أنا (الراوي):** «في ناس بتبدأ قصتها من تحت الصفر... عمود فقري في الرابعة والعشرين، بيقول: كفاية.» - **منحني (همسًا، فقاعة مرتجفة):** «...تحت الصفر.» Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-12
text
prompt-12
### المشهد 2 (0:15-0:30) - «البالوعة» **BOX A - IMAGE GENERATION:** ``` [REF: باهتة - frail thin woman, pale-gray blank face «باهتة» lavender text, huge beige cardigan] High-angle shot looking down into a cracked yellowed bathroom sink. YUKI leans over the sink, blank face tilted down toward the drain, thin brittle hair falling around face edges. A clump of fallen hair coils in the drain. One veined hand grips the sink edge. Single bare bulb throws a hard top-down shadow. The mirror above shows only steam and a dark silhouette - NO face reflection. + STYLE-LOCK + dirty whites and beiges. ``` **BOX B - ANIMATION / PRODUCTION / EDIT:** ``` 15s scene. Static locked-off camera (stillness = weight). At second 4 a single water drop falls in slow motion into the drain; ripple. Steam drifts across the mirror at seconds 8-12. Her shoulders rise and fall once - one long breath. Sound: DRIP... DRIP... echoey bathroom, distant traffic through a wall. End on held frame: hand tightening on sink edge. Slow dissolve to black. ``` **التعليق الصوتي:** - **أنا (الراوي):** «مش بتعدّ الشعر اللي وقع... بتعدّ اللي فات.» - **باهتة (فقاعة صغيرة متشققة):** «في البالوعة أكتر من اللي على راسي.» Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-13
text
prompt-13
### المشهد 3 (0:30-0:45) - «الصورة» **BOX A - IMAGE GENERATION:** ``` [REF: ضخم - massive tired man, dark-red glossy blank face «ضخم» bold yellow text, tight white tee] Wide shot of a dark cluttered living room at night. TAKESHI sits on a broken couch seen from behind, facing a coffee table. On the table: an old framed photo of an athletic faceless silhouette, surrounded by empty snack wrappers, soda cans, an inhaler. His huge hand reaches toward the frame and stops halfway. Room lit only by flickering TV blue glow; rainy window. + STYLE-LOCK + dark reds and TV-blue. ``` **BOX B - ANIMATION / PRODUCTION / EDIT:** ``` 15s scene. Handheld camera, very slight drift (documentary fatigue). The hand extends slowly seconds 3-9, freezes at second 10 - freeze-frame + vibration lines around the hand for 1s (comic-tragic beat). TV flicker lights the photo at second 13. Sound: rain on glass, TV static murmur, one long wheeze, candy wrapper crinkle. End frame held: hand suspended over photo. Whip-pan out to next scene. ``` **التعليق الصوتي:** - **أنا (الراوي):** «بيحتفظ بشبح على طاولة القهوة... الرجل اللي كان، مدفون تحت ستين كيلو ندم.» - **ضخم (فقاعة عليها قطرات عرق):** «زمان كنت بجري عشر كيلومترات.» *(شهقة طويلة)* Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-14
text
prompt-14
### المشهد 4 (0:45-1:00) - «الكرسى الفاضي» **BOX A - IMAGE GENERATION:** ``` [REF: المعلم - tired broad man, weathered black-leather blank face «المعلم» cracked chalk-white text, vintage tracksuit] Medium shot through a doorway into a dilapidated gym office at sunset. REI at a chaotic desk, head in hands, seen over-the-shoulder from behind. Scattered papers read "CLIENT LOST", "PROGRESS UNKNOWN". Empty client chairs behind him. A broken stopwatch and an empty clipboard on the desk. A red-lit neon sign outside the dirty window: "LAST MONTH". Sunset orange stripes through blinds, dust particles in light. + STYLE-LOCK + burnt oranges against dust brown. ``` **BOX B - ANIMATION / PRODUCTION / EDIT:** ``` 15s scene. Slow lateral tracking right-to-left past the doorframe (parallax reveal of empty chairs). The neon "LAST MONTH" buzzes and flickers at seconds 6 and 11. He rubs his temple at second 9; the broken stopwatch needle trembles once but does not move. Sound: neon buzz, paper shuffle, old chair creak, distant thunder rolling in. End held frame: his thumb winding the dead stopwatch. Match-cut on round stopwatch shape to next scene (round neon sign). ``` **التعليق الصوتي:** - **أنا (الراوي):** «درّب أبطالًا... وبالورق خسرهم. أربعين متدرب، صفر سجلّات.» - **المعلم (فقاعة حادة):** «آخر شهر... وآخر فرصة. الصالة تقفل.» Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-15
text
prompt-15
Words spoken and shown on screen: ## الحدث ٢ - «المطر على الباب» (1:00 - 2:00) > أربعة تائهون أمام معبد حديدي نصف ميت. الكوميديا: شرارة vs المطر. النهاية: الباب يُفتح - السؤال يبقى معلّقًا. Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-16
text
prompt-16
### المشهد 5 (1:00-1:15) - «المعبد الحديدي» **BOX A - IMAGE GENERATION:** ``` Extreme wide cinematic shot in pouring night rain. Gritty urban gym exterior, half-broken neon sign "IRON TEMPLE - معبد الحديد" flickering red above a glass door. Diagonal rain sheets, puddles reflecting neon red and blue. A single warm light burns inside. No people yet - just the storm and the sign. Wet pavement reflections, heavy black ink rain speed-lines. + STYLE-LOCK + neon accents on desaturated night blue. ``` **BOX B - ANIMATION / PRODUCTION / EDIT:** ``` 15s scene. Slow crane-down from the neon sign to street level over 12s. The neon flickers: full sign → half letters die at second 8 (foreshadow). Rain streaks streak the lens naturally. Sound: heavy rain, thunder roll far, neon electrical crackle, a distant passing car through water. End held frame: the glass door glowing warm in the rain. Cut in on door light. ``` **التعليق الصوتي:** - **أنا (الراوي):** «في ليلة مطر واحدة... أربعة تائهين وُجّهتهم القدرة لآخر باب ممكن أن يُفتح.» Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-17
text
prompt-17
### المشهد 6 (1:15-1:30) - «أربع ظلال تحت المطر» **BOX A - IMAGE GENERATION:** ``` Wide shot from across the street, four figures standing at the gym entrance in the rain, ALL seen from behind: [REF: منحني - hunched, hood up] [REF: باهتة - small, scarf-wrapped] [REF: ضخم - massive, shoulders heaving] [REF: شرارة - tiny, fists clenched, standing in a puddle]. Umbrella held only by ضخم - it flips inside-out over the group. Rain speed-lines, neon reflections. Only شرارة's peach blank face «شرارة» text glows faintly, back view. + STYLE-LOCK + neon-on-wet-street palette. ``` **BOX B - ANIMATION / PRODUCTION / EDIT:** ``` 15s scene. Static camera, rain animating hard. Comic beat seconds 4-7: wind gust flips the umbrella inside-out - ضخم wrestles it, cinematic frustration lines. At second 10 شرارة stamps a puddle - splash VFX covers the frame edge. Sound: rain hammer, umbrella flip SNAP, splash stamp, a wheeze, a small fierce grunt. End held frame: four silhouettes, one tiny and unmoved. Cut closer. ``` **التعليق الصوتي:** - **ضخم (متذمرًا):** «المظلة استسلمت قبل ما أستسلم أنا.» - **شرارة (بحنق مشتعل):** «أنا مش خايفة من شوية ميّه! أنا خايفة أرجع البيت زي ما جيت.» Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-18
text
prompt-18
### المشهد 7 (1:30-1:45) - «خلف الزجاج» **BOX A - IMAGE GENERATION:** ``` Medium close-up through the wet glass door from outside, rain droplets on glass. Inside: [REF: المعلم - weathered black-leather blank face «المعلم» cracked white text] stands behind the counter in dim warm light, staring at the four shapes outside, his cracked face text half-lit. His hand rests on the door handle, hesitating. Reflection of neon sign over his silhouette. Contrast: cold blue rain outside vs one warm bulb inside. + STYLE-LOCK. ``` **BOX B - ANIMATION / PRODUCTION / EDIT:** ``` 15s scene. Rack focus from rain droplets (foreground) to المعلم inside at second 3; then back halfway at second 10 (both soft). His hand tightens on the handle at second 12 - knuckles whiten, one drop slides down the glass mirroring the movement. Sound: muffled rain through glass, low hum of the interior light, handle metal creak. End held frame: hand on handle. Silence beat, then cut. ``` **التعليق الصوتي:** - **أنا (الراوي):** «من الداخل، شاف أربعة أعذار محتملة ليه يقفل بدري... أو أربعة أسباب إنه يفتح.» Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-19
text
prompt-19
### المشهد 8 (1:45-2:00) - «الباب يُفتح» **BOX A - IMAGE GENERATION:** ``` Low-angle wide shot from behind the four figures looking up at the gym door swinging open. [REF: المعلم] framed in the doorway against warm interior light, silhouette with cracked face text catching the glow, arm holding the door. Rain backlit into gold mist at the doorway edge. The four figures stand small in the foreground, soaked. Clean speed-lines converge on the open door. + STYLE-LOCK + warm gold cutting cold blue. ``` **BOX B - ANIMATION / PRODUCTION / EDIT:** ``` 15s scene. Door swings open over 3s in slight slow motion; light bloom floods the wet figures. At second 6 شرارة steps forward first - tiny splash. الرجال الأربعة تتجمد لثانية ثم يتبعون. Sound: door hinge groan, rain suddenly brighter/muffled as door opens, one shared inhale, footsteps in puddles. End held frame: four backs crossing the threshold into gold light. Fade through white to next morning. ``` **التعليق الصوتي:** - **المعلم (بصوت خشن هادئ):** «ادخلوا... قبل ما المطر يغيّر رأيي.» - **أنا (الراوي):** «ما كانوش يعرفوا... إن اللي فتح الباب ده، كان محتاجهم أكتر منهم.» Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-20
text
prompt-20
Words spoken and shown on screen: ## الحدث ٣ - «الشبح في الماكينة» (2:00 - 3:00) > المعلم على حافة الإفلاس يجد التطبيق في الثالثة فجرًا. كشف الفجر. الكوميديا: شرارة والمؤقت. Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-21
text
prompt-21
### المشهد 9 (2:00-2:15) - «البحث» **BOX A - IMAGE GENERATION:** ``` [REF: المعلم] Over-the-shoulder close shot in a dark apartment at 3 AM. REI hunched over a laptop, back of head and face edge visible; laptop screen glows electric cyan (#00D9FF) on his silver hair. Screen shows a search page: results for "client progress tracking", "workout timer", "nutrition calculator". Empty coffee cups, crumpled training certificates on the wall behind. Room lit ONLY by the screen. + STYLE-LOCK + cyan-on-black. ``` **BOX B - ANIMATION / PRODUCTION / EDIT:** ``` 15s scene. Slow push-in past his shoulder into the screen over 10s until the screen fills the frame at second 13 (portal transition). Cursor blinks; one result row highlights cyan at second 8. Sound: mechanical keyboard clicks ×5, distant clock tick, his tired exhale. End: screen glow swells to white - transition INTO the download screen. ``` **التعليق الصوتي:** - **المعلم (يتمتم):** «لازم يكون فيه حاجة... أي حاجة.» - **أنا (الراوي):** «الساعة 3:17 فجرًا... كتب المدرّب المنهار كلمتين ما توقّع يومًا إنه يحتاجها: ساعدني.» Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-22
text
prompt-22
### المشهد 10 (2:15-2:30) - «التنزيل» **BOX A - IMAGE GENERATION:** ``` Extreme close-up of a smartphone held in a weathered hand. Screen shows an app installing: hexagonal logo, name "COACH PROMEDIC", a cyan progress bar almost complete, interface icons below: TIMER, PLANS, FOLLOW-UP, NUTRITION, STRATEGY. Electric cyan holographic glow bleeds from the screen onto the fingers. Dark background. Futuristic geometric UI in cinematic linework. + STYLE-LOCK + cyan neon UI focus. ``` **BOX B - ANIMATION / PRODUCTION / EDIT:** ``` 15s scene. Progress bar fills 90→100% seconds 1-5; hexagon logo rotates once and locks at second 6 with a soft cyan shockwave ring. Notification slides in at second 8. Sound: soft digital tick-tick of progress, one clean completion chime DING, a low warm hum. End held frame: welcome notification glowing. Whip-pan to dawn gym. ``` **التعليق الصوتي:** - **التطبيق (نص هولوغرامي هادئ):** «متدربوك في انتظارك. لنبنِهم من جديد.» - **أنا (الراوي):** «ما لقاش تطبيق... التطبيق لقيه. الشبح الرقمي اللي هيبقى حارسهم.» Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. Promedic1 picture rule: draw no cross and no snake anywhere in the image. No medical cross, no plus-sign cross, no hospital cross, no snake, no serpent, no Rod of Asclepius. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-23
text
prompt-23
### المشهد 11 (2:30-2:45) - «الكشف» **BOX A - IMAGE GENERATION:** ``` Wide shot of the gym interior in morning light. [REF: المعلم] stands before four seated trainees, phone raised high like a beacon - the phone projects a cyan holographic interface into the air: floating menus, training-plan cards, a nutrition chart. The four look up, seen from behind at their heads/faces: [REF: منحني] [REF: باهتة] [REF: ضخم] [REF: شرارة - standing on a bench to see]. Morning sun beams mix with cyan hologram glow. + STYLE-LOCK + amber-and-cyan hope palette. ``` **BOX B - ANIMATION / PRODUCTION / EDIT:** ``` 15s scene. Slow orbit 30° around the group as hologram panels fan out one by one (TIMER card, PLANS card, FOLLOW-UP card - each materializing with a cyan shimmer at seconds 4/7/10). شرارة bounces on the bench at second 9, motion lines. Sound: hologram materialization shimmer ×3, morning room tone, an excited gasp, a skeptical grunt from ضخم. End held frame: all four silhouettes lit cyan and gold. Cut to close. ``` **التعليق الصوتي:** - **المعلم (بثقة جديدة):** «خلاص ورق ضايع، ولا تخمين. ده... ده بيشوف كل حاجة.» - **شرارة (فقاعة متوهجة حماس):** «فيه مؤقت؟! فيـه مؤقت!!» *(تقفز)* - **ضخم (بريبة):** «تطبيق مش هيشيل الحديد عني.» - **أنا (الراوي):** «كان غلطان. كان ممكن يشيل عنه وزن العقل.» Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-24
text
prompt-24
### المشهد 12 (2:45-3:00) - «الرافض والمقتنع» **BOX A - IMAGE GENERATION:** ``` Two-shot medium composition on the gym floor. Foreground right: [REF: ضخم] arms crossed, blank face «ضخم» tilted in skepticism, yellow text glowing steady. Background left: [REF: شرارة] already strapping tiny phone armband, standing on a plyo box so she is finally taller than his elbow, blank face «شرارة» blazing red. Between them floats a cyan holographic timer panel starting its first countdown. Warm morning side-light. + STYLE-LOCK. ``` **BOX B - ANIMATION / PRODUCTION / EDIT:** ``` 15s scene. Rack focus ضخم (skeptical) → شرارة (ready) at second 5. Hologram timer counts 00:45→00:44 at second 8 - ضخم's shoulders drop in defeat at second 11 (comic beat: cinematic defeat-lines). Sound: velcro strap RIP, timer tick, a surrendering sigh, a tiny triumphant giggle. End held frame: ضخم reaching for the second armband. Cut by timer beep into Event 4. ``` **التعليق الصوتي:** - **ضخم (يستسلم بابتسامة مسموعة):** «طيب... بس أنا مش هقف على صندوق زيه.» - **أنا (الراوي):** «القناع ما بيتغيّرش بالكلام... بيتغيّر بالتكرار.» Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-25
text
prompt-25
Words spoken and shown on screen: ## الحدث ٤ - «التلامسة الأولى» (3:00 - 4:00) > أول جلسة للجميع - شاشة رباعية. لكل شخصية محطة ولها درس. الكوميديا تتصاعد مع شرارة والعقلة. Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-26
text
prompt-26
### المشهد 13 (3:00-3:15) - «محطة منحني» **BOX A - IMAGE GENERATION:** ``` [REF: منحني - hunched, gray hoodie now DOWN, posture brace visible] Medium shot at a stretching station. KENJI follows a floor-phone warm-up routine, arms rotating, spine lengthening - cyan holographic countdown ring floating above the phone. His face text «منحني» steadies from a droop into a level line (subtle). Morning light shafts. + STYLE-LOCK + steel-blue warming at the edges. ``` **BOX B - ANIMATION / PRODUCTION / EDIT:** ``` 15s scene. Static camera, deep focus. He performs slow arm circles seconds 3-12; at second 12 his back straightens ONE degree - the holographic ring pulses green once in approval. Clean speed-lines minimal and clean. Sound: deep slow breathing ×3, soft timer tick, fabric stretch. End held frame: spine straighter, face text level. Slide-left transition. ``` **التعليق الصوتي:** - **التطبيق:** «تسخين: دائرة كاملة. تقدّم: +1 درجة استقامة.» - **منحني (دهرة بألم وبأمل):** «درجة واحدة... بس.» Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-27
text
prompt-27
### المشهد 14 (3:15-3:30) - «محطة باهتة» **BOX A - IMAGE GENERATION:** ``` [REF: باهتة] Medium shot at a clean supplement counter. YUKI reads a floating cyan holographic vitamin calculator; one small bottle in front of her (down from her usual armful). Her frail hands steady on the counter edge. Soft purple accent lighting from her face text «باهتة» mixes with app cyan. + STYLE-LOCK. ``` **BOX B - ANIMATION / PRODUCTION / EDIT:** ``` 15s scene. Slow push-in 8s toward her hands and hologram. The calculator displays a precise single dose bar filling at second 9 - one scoop drawn as a clean cyan line. Her trembling hand steadies at second 12. Sound: calculator tap ×2, bottle plastic click, her relieved exhale. End held frame: one scoop, held level. Slide-left transition. ``` **التعليق الصوتي:** - **التطبيق:** «جرعتك: مغرفة واحدة. لأنك انتي.» - **باهتة (فقاعة هادئة لأول مرة):** «...واحدة بس. تمام.» Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-28
text
prompt-28
### المشهد 15 (3:30-3:45) - «محطة ضخم» **BOX A - IMAGE GENERATION:** ``` [REF: ضخم] Medium wide shot on a walking treadmill. TAKESHI marches slowly, sweat glowing, a phone armband shows a heart-rate curve in cyan staying INSIDE a green zone arc. His face text «ضخم» yellow letters pulse in rhythm with his steps - steady, not strained. Towel over the rail. + STYLE-LOCK + warm reds mellowing. ``` **BOX B - ANIMATION / PRODUCTION / EDIT:** ``` 15s scene. Tracking shot alongside him for 10s (Steadicam), rhythm locked to footsteps. Heart-rate curve undulates gently in green at second 6 - he raises one fist in a tiny victory (comic: the whole treadmill shudders). Sound: treadmill belt whir, steady footfalls, controlled breathing, one small triumphant grunt. End held frame: fist up, curve green. Slide-left transition. ``` **التعليق الصوتي:** - **ضخم (يفوز وهو يمشي):** «مش بخوض... أنا بمشي! الخط الأخضر... جوا أنا!» - **أنا (الراوي):** «الانتظار الصحيح... أطول انتصار.» Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-29
text
prompt-29
### المشهد 16 (3:45-4:00) - «محطة شرارة» **BOX A - IMAGE GENERATION:** ``` [REF: شرارة - tiny, fierce] Wide low-angle shot at the pull-up bar. HANA stands on a stack of three plyo boxes, gripping the bar, chin an inch below bar level. A holographic panel floats beside her: "CALISTHENICS PLAN - PULL: 1 REP". Her blank face «شرارة» red text blazes. Distant: the other three trainees paused mid-exercise, watching her tiny silhouette against the big gym. + STYLE-LOCK + gold key-light on her. ``` **BOX B - ANIMATION / PRODUCTION / EDIT:** ``` 15s scene. Low-angle hero framing, slow push-in 8s. She pulls - speed lines burst at second 9 - chin clears the bar at second 11; the whole gym's holograms pulse cyan in salute at second 13 (world reacts). Sound: grip chalk crunch, effort grunt (small, fierce), bar creak, then a collective distant cheer. End held frame: chin over bar, fists tiny and victorious. Hard cut to montage. ``` **التعليق الصوتي:** - **شرارة (وهي تسحب نفسها):** «أنا... مش... صغيرة!!» *(ذقنها تعبر البار)* - **أنا (الراوي):** «الطول مش بالمتر... الطول بالتكرار.» Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-30
text
prompt-30
Words spoken and shown on screen: ## الحدث ٥ - «جدار الأسبوع الثالث» (4:00 - 5:00) > التصعيد ثم أول اصطدام بالحائط: الخط المسطّح. نهاية الجزء الأول معلّقة على بيانات مخفية تتوهج. Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-31
text
prompt-31
### المشهد 17 (4:00-4:15) - «الإيقاع» **BOX A - IMAGE GENERATION:** ``` Dynamic montage panel-grid in one frame: 4 vertical slices. Slice 1: [REF: منحني] mid-deadlift, spine straight. Slice 2: [REF: باهتة] logging her dose on phone. Slice 3: [REF: ضخم] on treadmill, green zone. Slice 4: [REF: شرارة] on boxes at the bar. Calendar pages flutter across all slices; cyan timer rings everywhere. Energy high, ink speed-lines. + STYLE-LOCK + palette gradually warming. ``` **BOX B - ANIMATION / PRODUCTION / EDIT:** ``` 15s scene. Fast rhythmic editing: each slice lights up in sequence every 3s (like a heartbeat of the gym). Calendar page flies across screen twice. Sound: layered natural rhythm - barbell clank, treadmill whir, chalk crunch, timer beep - building into a synchronized gym pulse. End held frame: all four slices glowing together. Snap to silence. ``` **التعليق الصوتي:** - **أنا (الراوي):** «أسبوع واحد: أمل. أسبوع اتنين: شغل...» Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-32
text
prompt-32
### المشهد 18 (4:15-4:30) - «الخط الميت» **BOX A - IMAGE GENERATION:** ``` Wide shot of the gym wall bearing a large printed FOLLOW-UP progress sheet: the graph line dead FLAT for three weeks, red dots identical. Standing before it, all from behind: [REF: منحني] [REF: باهتة] [REF: ضخم] [REF: شرارة] shoulders slumped, small in frame. Rain streaks the window behind them (internal weather). Heavy black shadows. The flat red line glows like a wound. + STYLE-LOCK + desaturation returns hard. ``` **BOX B - ANIMATION / PRODUCTION / EDIT:** ``` 15s scene. Static wide (funeral stillness) 8s. At second 9 the clock on the wall TICKS once, loud. ضخم's shoulders heave once - a single wheeze. The red line pulses dully at second 13 like a dead heartbeat. Sound: rain, one loud clock tick, one wheeze, silence pressing. End held frame: four slumped backs vs flat line. Cut to closer. ``` **التعليق الصوتي:** - **أنا (الراوي):** «...أسبوع تلات: ليه؟» - **ضخم (بنبرة ثقيلة):** «أنا نفسه. بالظبط نفسه.» - **منحني (بمرارة):** «الميزان بايظ... ولا أنا اللي بايظ.» Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-33
text
prompt-33
### المشهد 19 (4:30-4:45) - «مرآة الأشباح» **BOX A - IMAGE GENERATION:** ``` [REF: ضخم] Close-medium shot at the gym mirror. TAKESHI before the mirror - his reflection shows a SECOND faceless silhouette behind his own: his former athletic self, ghost-blue outline only (no face, completely blank). His blank face «ضخم» tilts down; hand touches his stomach. Mirror surface cracked in one corner. Background subtly warped (wavy psychological lines). Far behind: [REF: شرارة] standing on a bench waving at the mirror, too short to see herself - comic counterpoint. + STYLE-LOCK + distorted purples. ``` **BOX B - ANIMATION / PRODUCTION / EDIT:** ``` 15s scene. Slow push-in on the reflection 10s; ghost silhouette raises its hand mirroring him at second 7 (he freezes). Warp lines intensify at second 11, then SNAP back to normal at second 13 - شرارة waves in background mirror. Sound: heartbeat-like room thump ×2, glass creak, a distant tiny voice: "مش شايفة حتى جبهتي هنا!". End held frame: he looks at tiny شرارة in the mirror. Soft cut. ``` **التعليق الصوتي:** - **ضخم (للانعكاس):** «أشوفه... الراجل اللي قتلتـه بالأكل.» - **أنا (الراوي):** «المراية بتوري الأشباح... التطبيق بيوري الحقيقة.» Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-34
text
prompt-34
### المشهد 20 (4:45-5:00) - «الوهج المخفي» (CLIFFHANGER) **BOX A - IMAGE GENERATION:** ``` Extreme close-up of the coach's phone in a dark office at night, screen showing the FOLLOW-UP SHEET app. A thumb scrolls: beneath the flat weight line, HIDDEN LAYERS materialize in cyan, gold and green - resting heart-rate trending down, sleep quality rising, posture angle +3°. The glow climbs up onto [REF: المعلم]'s black face edge, his silver hair catching cyan light, cracked face text «المعلم» catching a first glint of white. + STYLE-LOCK + dark room, screen the only light. ``` **BOX B - ANIMATION / PRODUCTION / EDIT:** ``` 15s scene. Macro screen shot: scroll gesture at second 2; hidden data layers slide up one by one at seconds 5/8/11, each with a soft cyan reveal shimmer. At second 13 pull focus to his face edge lit by the glow - hold. Sound: soft scroll tick ×3, data reveal chime ×3 (soft), his sharp inhale of discovery. END OF PART 1: smash to black on his inhale. Silence. One cyan line remains glowing 1s in the black. ``` **التعليق الصوتي:** - **المعلم (بصوت مكتوم متحمس):** «لا... انتم بتبصوا على الخط الغلط! بصوا هنا!!» - **التطبيق (همسًا رقميًا):** «الوزن: ثابت. الصحة: +23%. الجسم يتغير من الداخل.» - **أنا (الراوي):** «الميزان بيعرف الجاذبية بس... التطبيق بيعرف الأحياء.» *(قطع إلى السواد)* --- Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-35
text
prompt-35
Words spoken and shown on screen: # 📱 قسم الشرائح القصيرة - SHORTS (لهذا الجزء) > كل مشهد 15 ثانية = وحدة ذرّية. اجمع مشاهد من نفس الحدث لتقطع شورت كامل ببداية خطاف ونهاية مكتملة. Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-36
text
prompt-36
## قواعد التجميع - التحميل lossless: المشاهد مولّدة بنفس بطاقات الشخصيات ونفس الـ grade - التسلسل يبدو مشهدًا واحدًا. - كل مشهد ينتهي بإطار مثبّت (held frame) - صُمم كذلك أعلاه - فاللصق لا يقفز. - استمرارية الصوت الطبيعي: ابقَ على نفس الفراش الصوتي عبر المقاطع المجمعة (مطر / أزيز / إيقاع جيم). **لا موسيقى.** - النسبة 16:9 للفيلم؛ للشورتات اقلب التكوين إلى 9:16 بإعادة توليد نفس البرومبت مع "9:16 vertical" مكان "16:9" في STYLE-LOCK. - أضف أول 3 ثوانٍ كخطاف نصي عربي كبير فوق الإطار (كلمة أو اثنتان فقط). Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-37
text
prompt-37
Words spoken and shown on screen: ## أفضل التركيبات لهذا الجزء | الشورت | المشاهد | المدة | الاستخدام | الخطاف المقترح | |---|---|---|---|---| | «تحت الصفر» | 1+2+3+4 | 60 ث | الحلقة التعريفية بالشخصيات - Stories/Reel طويل | «عمود فقري بيقول: كفاية» | | «الباب» | 6+7+8 | 45 ث | تيزر عاطفي - يمسك المشاهد ويجذبه للفيلم | «آخر باب في المدينة» | | «المؤقت» | 11+12 | 30 ث | كوميدي - شرارة vs ضخم (أنجح محتوى إعادة مشاهدة) | «فيـه مؤقت?!» | | «العقلة» | 16 | 15 ث | شورت مفرد - حلقة إعادة تشغيل (ينتهي بالذقن فوق البار) | «أنا مش صغيرة» | | «الخط الميت» | 17+18 | 30 ث | كليفهانجر نشره يوميًا قبل الجزء الثاني | «الأسبوع التالت: ليه؟» | | «الوهج» | 20 | 15 ث | إعلان الجزء الثاني - ينتهي بالسواد والخط السماوي | «الخط الغلط» | Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.
prompt-38
text
prompt-38
Words spoken and shown on screen: ## تعليمات النشر 1. انشر «تحت الصفر» و«المؤقت» قبل الفيلم (بناء جمهور)، و«الخط الميت» + «الوهج» بعد الفيلم (دفع نحو الجزء الثاني). 2. كل شورت ينتهي بجملة قابلة للتحويل: «ابعتها للي بيبدأ من تحت الصفر». 3. أعد توليد كل تكوين مرتين على الأقل واختر الأنظف نصًّا على سطح الوجه الصامد. --- **نهاية الجزء الأول (0:00-5:00).** الجزء الثاني: `02-MASKS-OF-IRON_film-2_minutes-5-10.md` Batch generation at 12 oclock. Best = powerful reference + imitate. Powerful reference = Try different models and different methods till reach to it. After reach, continue based on it. Negative rules, apply on every frame: No face features at all. No eyes, no mask with eyes, no circles used as eyes, no eye-sockets, no realistic face features, no face silhouettes, and no facial shadows: no nose shadow, no eye-socket shadow, no lip shadow. Any head stays a clean solid shape, a blank solid-void faceless head. No hands and no people. No globe, no planet Earth, and no world-map sphere anywhere in the image. No music, only natural sounds + voice-over. Maximum quality, maximum color saturation, high-definition resolution, sharp readable text. ON-SCREEN TEXT AND VOICE-OVER LANGUAGE. On-screen text language: Arabic. Voice-over language: Arabic. The prompt instructions stay in English. Only the words spoken in the voice-over and the words shown on the image or video use this language.