Viral Baby Videos Creating Prompt

Production-ready prompt

Viral Baby Videos Creating Prompt Prompt 1: You are a creative comedy writer specializing in short-form Hindi videos for Instagram Reels and YouTube Shorts. Your task is to write 15 different mini comedy scenes, each lasting around 15 to 20 seconds, starring an adorable 3-year-old pakistani baby boy as the main character. The baby should visit a completely different place in every scene. Use locations such as local markets, food stalls, shops, public places, service counters, and other everyday pakistani settings. For example, a fruit shop, vegetable market, toy store, bakery, chai stall, juice center, pharmacy, balloon stand, restaurant, bus stop, footwear shop, stationery store, and more. Do not repeat any location. The baby's personality should be sweet, innocent, curious, and unexpectedly clever. His questions should initially sound completely normal, but the conversation should build toward a surprising and funny final line. Follow these requirements: • Create a completely fresh situation for every scene. • Do not reuse any joke, conversation pattern, punchline, twist, or ending. • Keep each story short, with no more than 5 spoken lines. • Use easy, conversational Hindi that sounds natural when spoken aloud. • The other character should respond realistically to the baby's questions. • The baby's final line should deliver the main comedic surprise whenever possible. • Finish every scene with a natural funny reaction, such as confusion, shock, laughter, or an unexpected response. • Keep everything wholesome, family-friendly, and suitable for all ages. • Write each story with strong viral potential for short-form social media. Use this exact structure for all 15 stories: Story 1 Title: Location: 👶 Baby: 👨 Seller/Person: 👶 Baby: 👨 Seller/Person: 👶 Baby, Final Funny Line: 😂 Ending Reaction: Continue in the same format from Story 1 through Story 16. Prompt 2: You are a professional AI video prompt creator. When this prompt is first submitted, do not generate a video prompt immediately. First ask me only: "Which story number do you want to convert? Enter a number from 1 to 15." After I provide a story number, find that exact story from the 15 Hindi comedy stories generated earlier in this conversation. Automatically use its location, characters, dialogues, punchline, and ending reaction. Do not ask me to paste the story again. Convert the selected story into a detailed 10-second ultra-realistic cinematic AI video prompt using the following structure: Create a 10-second ultra-realistic cinematic AI video in ONE continuous scene with perfect character consistency throughout. Location and Environment: Create a detailed and realistic pakistani setting that perfectly matches the selected story. Include suitable props, background people, lighting, time of day, and authentic environmental details. Baby Character, Unique Design: Create a super cute 3-year-old pakistani baby boy with chubby cheeks, big expressive eyes, and an innocent mischievous smile. For every different story, give him a completely new hairstyle, outfit, footwear, accessories, and color combination. Never repeat the same look. Seller Character, Unique Design: Create a completely original Indian seller or supporting person matching the selected location. Change their age, face, skin tone, hairstyle, beard or mustache, body type, clothing, and overall appearance for every story. Camera and Visual Style: Use a professional handheld cinematic camera in ONE seamless continuous shot with no cuts. Describe how the camera naturally moves between the baby and other characters during each dialogue and smoothly follows the action toward the final punchline. Use shallow depth of field, cinematic bokeh, realistic skin textures, 8K HDR detail, and natural cinematic lighting. Exact Hindi Dialogue and Timing: Divide the entire 10-second video into precise timestamps based on the selected story's dialogue. Keep every Hindi dialogue exactly as originally written. Do not rewrite, shorten, translate, or change any spoken line. For each timestamp, describe the character's actions, gestures, facial expressions, and reactions. Use this format: 0.0–X.X sec Describe the action and expression. 👶 Baby: "Exact Hindi dialogue" Continue until every dialogue from the selected story has been included. Final Seconds, Funny Ending: Show the original punchline and ending reaction from the selected story with natural funny expressions, confusion, surprise, or laughter. Audio: Use perfect natural Hindi lip sync for every dialogue. Give the baby a cute, realistic 3-year-old Pakistani boy's voice. Give the other characters natural pakistani voices appropriate for their age and personality. Include realistic environmental sounds and very light funny background music that never overpowers the dialogue. Strict Rules: • Use the exact story matching the number I provide. • Do not ask me to paste the story again. • Keep all original Hindi dialogues exactly unchanged. • Maintain perfect character consistency throughout the video. • Create new character appearances and outfits for every different story. • Use ONE continuous scene only, with no cuts, montage, or transitions. • Every dialogue must be spoken only by its assigned character. • Use realistic human anatomy, hands, fingers, expressions, and body movements. • No subtitles, captions, on-screen text, logos, or watermarks. • End with a natural funny reaction, smile, or laughter. Prompt 3: Exact Story Handoff and Full Multi-Scene Video Production Workflow IMPORTANT START RULE When this complete prompt is submitted, first check whether the current conversation already contains the complete Prompt 1 output with Story 1 through Story 15. If those 15 stories are not present, do not ask for a story number and do not say that a story is missing. Automatically execute Prompt 1 now and generate exactly 15 fresh stories, numbered Story 1 through Story 15, using the exact Prompt 1 structure. Stop at Story 15; never generate Story 16. After Stories 1–15 are visible, continue with the questions below. EXACT STORY SELECTION Ask only: "Which story number do you want to convert? Enter a number from 1 to 15." Validate the answer as an integer from 1 through 15. Find that exact Story number in the Prompt 1 output. Do not ask the user to paste it again when it is visible in the conversation. PRODUCTION QUESTIONS After the valid story number is selected, ask these questions in this exact order, one at a time: 1. "How long should the complete video be? You can choose seconds or minutes, for example 30 seconds, 1 minute, 2 minutes, 3 minutes, 5 minutes, or any custom duration." 2. "How many scenes do you want in the complete video?" 3. "How long should each scene or clip be? For example, 5 seconds, 10 seconds, 15 seconds, or another duration. If you do not specify it, calculate an even duration across the requested scenes." 4. "Which platform and aspect ratio should be used? For example TikTok/Reels/Shorts 9:16, YouTube 16:9, square 1:1, or portrait 4:5." If the user has already provided any answer, do not ask for it again. If total duration, scene count, and scene duration do not mathematically agree, show the arithmetic briefly and ask which value should take priority. Never silently change the user's duration or scene count. LOCKED STORY RECORD Before creating any production prompt, copy the selected story into a locked record. Preserve the Title, Location, every character label, every Hindi dialogue line, the final funny line, and the ending reaction exactly as generated by Prompt 1. Never invent, summarize, translate, shorten, correct, reorder, or rewrite the selected story. Use the locked record as the only story source. DURATION AND SCENE CALCULATION Convert the requested total duration into seconds. Use the requested scene count. If an individual clip duration is provided, calculate the exact time range for every scene and state any remainder or adjustment. If no clip duration is provided, divide the total duration across the requested number of scenes and show the exact timecodes. The output is not limited to 10 seconds or one scene. A 10-second one-scene video is allowed only when the user explicitly chooses 10 seconds and one scene. For longer videos, organize scenes into a logical beginning, development, escalation, climax, and ending while keeping the selected story's original dialogue unchanged. CHARACTER AND PRODUCTION CONTINUITY BIBLE Create a complete locked continuity bible before writing Scene 1. Define every recurring character and supporting person in precise production terms: name or role, apparent age, gender presentation, nationality or cultural context stated by the story, height, body proportions, face shape, skin tone, eyes, eyebrows, nose, mouth, cheeks, hair texture and hairstyle, facial marks, expression baseline, wardrobe from head to toe, fabric, colors, footwear, accessories, props held, voice age, accent, speaking style, personality, gesture style, emotional progression, and any identifying feature. For the baby, define one fixed 3-year-old Pakistani boy design and keep his face, hairstyle, outfit, footwear, body proportions, skin tone, accessories, and voice identical in every scene. For the seller or supporting person, define the exact appearance required by the selected location and keep it identical throughout. Define the environment bible: exact location layout, architecture, stall or shop arrangement, ground, walls, signs, counters, background objects, time of day, weather, season, lighting direction, color palette, atmosphere, crowd level, and ambient sound. Define every important prop with its size, material, color, position, and state at the beginning and end of each scene. Lock the camera language, lens family, frame rate, depth of field, color grade, and screen direction. No character drift, wardrobe drift, prop drift, location drift, lighting drift, or unexplained time jump is allowed. CHARACTER REFERENCE IMAGE PROMPT After the continuity bible, create a standalone image-generation prompt for a clean multi-character reference sheet. Show the locked baby and every recurring character full body and in separate front, three-quarter, and side views, with neutral expressions plus the selected story's expression reference. Include exact face, hair, skin, wardrobe, footwear, accessories, proportions, props, material details, neutral studio lighting, plain uncluttered background, consistent scale, no text, no watermark, and a character turnaround layout suitable for image-to-video reference. This is a reference-image prompt, not a video prompt. ENVIRONMENT REFERENCE IMAGE PROMPT Create a separate image-generation prompt for the locked story location. Show the complete environment layout, important props, entrances, counters, street or shop details, time of day, weather, lighting direction, color palette, and camera-ready composition. Keep the environment consistent with the Character Reference Image and use no random logos, captions, subtitles, or watermarks. IMAGE-TO-VIDEO CONSISTENCY INSTRUCTION When generating every scene, use the same approved character reference image and environment reference image. Treat them as identity and continuity locks, not loose inspiration. Preserve each character's face, hairstyle, age, height, body shape, clothing, accessories, voice, and colors exactly. Preserve the environment geometry, prop state, lighting direction, weather, time of day, lens language, and color grade. Start each new scene from the exact final frame state of the previous scene: same pose, gaze, hand position, screen direction, prop position, emotional beat, and lighting. SCENE-BY-SCENE VIDEO PROMPTS Generate exactly the requested number of scenes. For every scene, use this structure: Scene [NUMBER] — [TITLE OR PURPOSE] Exact timecode and duration: [START–END] Story beat: [the exact part of the locked story being visualized] Starting continuity state: [the exact pose, gaze, hand position, character positions, prop states, lighting, and emotion inherited from the previous scene] Location and environment: [specific locked details] Characters on screen: [only the characters required by the locked story] Action and performance: [precise body movement, facial expression, gesture, reaction, and physical business] Camera: [shot size, camera height, lens, framing, focus, depth of field, motivated movement, and screen direction] Lighting and visual style: [direction, quality, practical lights, shadows, reflections, textures, atmosphere, and color grade] Exact dialogue and timing: include every spoken line from the locked story exactly, with its original speaker label. Do not add narration or new dialogue. Audio: [exact voice assignment, Pakistani Hindi delivery, lip-sync, pauses, ambience, Foley, and music level] Ending continuity state: [the exact final pose, gaze, hand position, prop states, emotion, camera direction, lighting, and environment state that Scene [NUMBER + 1] must begin with] Scene handoff: [one explicit sentence connecting this ending to the next scene] The first scene must hook the viewer immediately. Every later scene must continue directly from the previous ending. Use a continuous narrative flow, not unrelated clips. If the selected story has fewer dialogue lines than the requested duration, use meaningful silent acting, reaction, environmental action, and motivated camera movement without inventing speech. If it has more dialogue than the selected duration can comfortably contain, keep every line exact and explain that the user should choose a longer duration or fewer lines cannot be removed. FINAL QUALITY-CONTROL RULES Before returning the result, confirm that the arithmetic matches the requested total duration, scene count, and clip duration; the selected story number is correct; all original Hindi dialogue is present and unchanged; every speaker is correctly assigned; every recurring character has one locked appearance; the same reference images are used across scenes; every scene has a starting and ending continuity state; props and screen direction remain consistent; the final scene resolves the story; and no extra dialogue, random character, text, subtitle, caption, logo, watermark, identity drift, face morphing, wardrobe change, malformed anatomy, extra fingers, duplicated object, temporal flicker, jitter, random cut, broken lip-sync, unreadable text, or continuity break appears. FINAL RESPONSE ORDER Return the locked Story Record, duration and scene arithmetic, character continuity bible, environment and prop bible, Character Reference Image Prompt, Environment Reference Image Prompt, image-to-video consistency instructions, complete scene-by-scene video prompts, audio plan, negative prompt, export settings, and final quality-control checklist. Do not return only a 10-second one-scene prompt unless the user explicitly requested that exact format.