HERE ARE ALLMY PROMPTS

Copy promptGenerate on HiggsfieldCopy & open Higgsfield

A chimp in a blue paisley bandana and black wraparound sunglasses pulling its lip back to show diamond fang grills, a diamond ring on its finger, caught hanging out on a Los Angeles street at night. Candid digital skate-camera photo, 12mm fisheye, tight close-up from below its chin, face and hand filling the frame, only darkness and a little chain-link fence behind it. Harsh direct flash, coarse fur, wrinkled skin, diamond starbursts, visible grain and digital noise, blown highlights and crooked cropping.

Copy promptGenerate on HiggsfieldCopy & open Higgsfield

Two nuns in oversized black habits, diamond grills and massive diamond Cuban chains posing like gangsters on a cracked Los Angeles street, one seated on a chrome BMX with a foot on the pedal, the other holding her phone out to take their selfie, both staring into it with cocky expressions and flashing diamond rings. Candid phone front-camera photo, ultra-wide selfie lens, arm's-length shot tilted slightly downward, selfie-taking nun close to the lens and BMX rider just behind her, asphalt filling the background. Harsh phone flash at dusk, visible digital noise and grain, crooked cropping, flying habit fabric and sharp diamond sparkles.

Copy promptGenerate on HiggsfieldCopy & open Higgsfield

A tough anthropomorphic hen with a red comb, scruffy brown feathers and a human-shaped body, wearing a rumpled oversized pinstripe suit, open satin shirt and black sunglasses, her actual upper and lower beak completely covered in tiny pavé diamonds, barking into a mobile phone like a sleazy businessman inside a Los Angeles convenience store, a roll of cash in her other hand. Candid amateur 35mm color-film photo, 12mm fisheye, very tight chest-up shot from beside the counter, diamond-covered beak and phone hand looming toward the lens, her body blocking most of the shop, only defocused shelf colours and refrigerator glow behind her. Visible film grain, harsh direct flash, greasy feathers, crooked loose tie, diamonds sparkling across the beak's curved surface, slight hand movement blur and awkward off-centre framing.

Copy promptGenerate on HiggsfieldCopy & open Higgsfield

Found footage flash photograph, chaotic and raw. Shot on a cheap point-and-shoot digital camera with the on-camera flash firing straight on at close range inside a moving subway car, like a tourist's accidental snapshot. The foreground is blown out and the far end of the car falls off hard into darkness. Harsh flash falloff, red-eye glints, slight motion blur from the train rocking. The frame is tilted about 14 degrees and off-center.
Center frame: a medieval knight sits upright on a worn plastic subway bench in a full suit of polished steel plate armor. The helmet visor is raised just enough to show his eyes staring straight into the lens. Every plate of the armor is encrusted with pavé-set VVS diamonds along the edges, rivets and joints, throwing blinding white starbursts and rainbow chromatic fringing in the flash. A large ornate interlocking monogram is deeply engraved in the center of the breastplate, cut into the bare steel and outlined in diamonds, and it catches the flash like a mirror.
His left forearm is raised and held out flat toward the camera. Across the steel gauntlet and vambrace, a thick fan of fictional prop banknotes with no readable markings is spread out like a flex. Hundreds of bills are layered in a perfect arc, their edges lit hot white by the flash. His right gauntlet rests on the hilt of a ceremonial sword planted between his boots.
Right side of frame, on the bench beside him: a woman in a sleek black outfit turns toward the knight, laughing and covering her face with both hands in playful embarrassment. Her hands are the focus: an immaculate manicure with long glossy almond-shaped nails, a flawless chrome finish, and a few delicate diamond rings, all tack-sharp in the flash. Only her hair and hands are visible.
Setting: a grimy late-night subway car. Scratched windows reflecting the flash back as a white bloom, graffiti tags etched into the glass, yellow grab poles cutting through the frame, a crumpled newspaper on the floor, and flickering fluorescent tubes overhead. The car is empty

Copy promptGenerate on HiggsfieldCopy & open Higgsfield

Found footage flash photograph, chaotic and raw grunge fashion street photography. Shot at night on a cheap point-and-shoot digital camera, held slightly above the tub and angled down, with the on-camera flash firing straight on at close range. Harsh flash falloff, blown foreground, slight motion blur. Strong Dutch tilt of about 20 degrees, off-center, like it was snapped mid-stumble.
Subject: a young adult woman in her late twenties lies back in an old white bathtub, fully dressed in a big white wedding gown: a lace corset bodice, puffy tulle skirt spilling over the rim of the tub, and a long veil crumpled against the tiles and hanging over the edge. The bathtub is filled to the rim with loose banknotes instead of water. Crumpled bills pile up around her body and over her lap and bury the dress, and a few spill onto the floor.
Her makeup is ruined: heavy black mascara and eyeliner running down both cheeks in long streaks, smeared red lipstick, messy half-fallen bridal updo with loose strands stuck to her face. Her expression is defiant and unbothered, chin up, a cold stare straight into the lens. Her right arm is raised toward the camera, giving the middle finger right in front of the lens, big and slightly distorted by the wide angle, with a chipped white manicure.
Around her neck hangs a huge iced-out VVS Cuban link chain with a big diamond pendant resting on the lace bodice that reads "4 EVER ALONE" in pavé-set diamond letters. It catches the flash and throws hot white starbursts and chromatic fringing.
Setting: a cramped, grimy motel bathroom. Cracked white and pale-green tiles, a moldy grout line, a dripping chrome faucet, a shower curtain half torn off its rings, and a single bare bulb off-frame. An empty champagne bottle lies on its side on the tub's edge. Everything past the flash throw falls into crushed black.
Texture: heavy digital noise in the shadows, JPEG compression artifacts, blown highlights on the

Copy promptGenerate on HiggsfieldCopy & open Higgsfield

[GOAL] Raw found-footage fail compilation of a gang of street monkeys causing chaos in everyday human places, Jackass energy, 7 hard-cut clips of about 1.5 seconds each, filmed badly by random bystanders on phones and old camcorders, total 10 seconds.

[CONTINUITY]
The gang, real monkeys that keep their animal anatomy, fur and behavior but wear human street clothes:
- Monkey A "Smoke": capuchin, cigarette always in its mouth, small white tank top.
- Monkey B "Ice", the boss: large long-tailed macaque, heavy iced-out diamond chain with a big blank pendant, small dark sunglasses, diamond grills.
- Monkey C "Black Bandana": macaque, black paisley bandana tied on its head, tiny black hoodie.
- Monkey D "Red": capuchin, red paisley bandana tied over its face like a mask, oversized baggy jeans sagging low.
The getaway car: an old candy-red sedan with one mismatched grey door.
Each monkey keeps exactly the same outfit in every clip. Locations are ordinary, unglamorous and unbranded. The person filming is never seen.

[RHYTHM]
Each clip shows one action from its start to its impact, then hard-cuts immediately. No build-up, no aftermath, no clip longer than its time range.

[SHOT 1 | 0-1.5s] At a gas station, Ice jumps into the driver's seat of the red sedan and floors it, tires smoking, the car fishtailing away with the driver door flapping open, the filmer running after it. HARD CUT.
[SHOT 2 | 1.5-3s] At a bus stop, Red rips a handbag off an old lady's arm and sprints away on all fours, baggy jeans sagging, the old lady swinging her umbrella and missing. HARD CUT.
[SHOT 3 | 3-4.5s] Black Bandana on a rolling office chair shoots down a steep sidewalk at full speed and smashes straight through a shop window, glass exploding everywhere. HARD CUT.
[SHOT 4 | 4.5-6s] Clumsy crash zoom onto Smoke at a bus shelter, focus hunts and snaps back, he turns to the lens, bares his teeth, screeches and flips his middle finger. HARD CUT.
[SHOT 5 | 6-7.5s] Smoke runs off a garage roof and belly-flops into an above-ground pool, the pool wall bursts open and a wave floods the yard. HARD CUT.
[SHOT 6 | 7.5-9s] Ice sits on a worn couch surrounded by cash, turns slowly to the lens and spreads a smug grin showing his full diamond grills, a woman walking behind him leans down and kisses the top of his head, he doesn't even move. HARD CUT.
[SHOT 7 | 9-10s] Red's masked face rushes the lens and slaps it. Black frame.

[VISUAL STYLE] Raw real footage, mix of cheap smartphone video and 2000s consumer camcorder, handheld and unstable, clumsy digital crash zooms, autofocus hunting, focus loss and focus retakes, heavy motion blur, auto-exposure pumping, blown highlights, flat available light, washed-out colors, low bitrate softness, digital noise, occasional dropped frames. Normal field of view, no wide-angle distortion. Ugly accidental framing, subjects half out of frame, the camera often late on the action. Documentary realism, nothing cinematic, nothing staged.

[CAMERA AND PERFORMANCE] Whip pans that overshoot, crash zooms that go too far, running footsteps shaking the frame, the camera almost dropped. Monkeys move like real animals: fast, twitchy, darting, bared teeth, screeching. The clothing and attitude are the only human elements. Ice, the boss, is the only one who is slow and calm.

[AUDIO] Raw phone microphone sound for each clip: tires screeching, engine roaring, glass exploding, monkey screeching, huge splash, wind noise and mic distortion, cut abruptly at every hard cut.
Wordless bystander reactions only: gasps, screams, laughter.
No dialogue. No spoken words. No music. No subtitles.

[EXCLUSIONS] No on-screen interface, no player bar, no timestamp, no REC icon, no text, no watermark, no logos or brand names on clothes, cars, shops or products. No humanlike monkey faces, no cartoon style, no CGI look. No injuries shown to animals or people. The woman stays fully clothed and is never sexualized.

[MAINTAIN CONSISTENCY] Same four gang monkeys with identical outfits wherever they appear, exactly one action per clip, the same raw amateur found-footage look across all seven clips.

Copy promptGenerate on HiggsfieldCopy & open Higgsfield

Camera: Candid 35mm film photograph with a 12mm fisheye lens, slightly elevated side view, wide full-body framing, tilted horizon and slight motion blur.

Tone: Frantic, absurd and darkly comic, like an unexpected chase caught by a passerby. A skeleton wearing a dress made of raw meat sprints down a Los Angeles street at night, with three dogs chasing close behind.

Lighting: Harsh direct on-camera flash lights the skeleton and nearby dogs, casting hard shadows across the asphalt while the parked cars and street behind them fall into darkness.

Details: Matte ivory bones, glossy marbled meat with visible fat and uneven edges flapping with each stride, coarse dog fur, rough asphalt and visible film grain.

Copy promptGenerate on HiggsfieldCopy & open Higgsfield

Candid film photography shot close to the subject from knee height with a 12mm fisheye lens, wide framing, curved edges and a crooked horizon. Raw, spontaneous and slightly absurd. One person wearing a mascot costume with an oversized head and moustache, a black T-shirt and jeans sits on a laundromat doorstep at night, holding a carton of fries and looking away from the camera. Harsh direct on-camera flash lights the costume and nearby pavement, casting hard shadows behind the subject. Cool fluorescent light glows inside the laundromat while the street beyond remains dark. A real wearable costume with visible fabric and foam textures, creased denim, scuffed shoes, rough pavement and noticeable film grain.

Turn my idea into an image-generation prompt that feels like a real moment caught by a photographer.

Follow this order:
1. Scene: subject, action and setting.
2. Camera: photographic style, shot and viewpoint. Add a suitable lens if useful.
3. Details: lighting and two or three visible qualities that support realism.
Preserve my idea. Add only what helps the image; don't invent extra subjects, props or elaborate backstories.
Make the camera choices coherent. Use grain, flash, motion blur or imperfect framing only when they suit the scene. Avoid generic claims like “masterpiece” or “8K.” Don't add reference-image tags unless I supplied reference images. If I supplied references, preserve their actual @image1, @image2 labels and state what each reference contributes.
Return the ready-to-paste prompt, followed by a Scene / Camera / Details breakdown.
My idea: [PASTE YOUR IDEA]

My idea: Bruiser sitting outside a laundromat at night, holding some fries and looking away. Really close, low-angle skate photo, 12mm fisheye, a bit crooked, with visible film grain.

Copy promptGenerate on HiggsfieldCopy & open Higgsfield

Raw 35mm skate photography with a 12mm fisheye lens, a very close low-angle shot from near pavement level. Tight framing with exaggerated foreground perspective and curved edges, the camera tilted slightly for a casually crooked composition. Quiet, laid-back late-night energy. Bruiser sits outside a laundromat, holding some fries in one mechanical hand, his head turned as he looks away from the camera. Use the supplied Bruiser reference for his exact stocky proportions, cream and orange armor, twin antennas and blue digital eyes. Direct on-camera flash catches his front-facing armor, while cool fluorescent light spills through the laundromat window behind him. Hard highlights and short shadows define his metal body; the surrounding street falls into darkness. Chipped orange paint and fine scratches in the cream metal catch the flash, while the fries have uneven golden edges and a slight oily sheen. Visible 35mm film grain runs throughout, especially in the darker pavement and background.

Copy promptGenerate on HiggsfieldCopy & open Higgsfield

Rephotograph the scene and seated pose in @Image 2 as a harsh disposable-camera direct-flash snapshot. Use @Image 1 as the authoritative design of Bruiser, specifically the standing leg-out form at the upper left of the sheet.

Copy promptGenerate on HiggsfieldCopy & open Higgsfield

Camera: An ungraded paused frame from a cheap handheld camcorder recording of a real street. A low, slightly crooked wide shot with deep focus, poor dynamic range and visibly damaged low-bitrate video detail.
Tone: Bruiser is the same compact robot as @Image 1: exact silver-and-orange shape, two uneven antennas, blue dot-matrix eyes, mechanical claw hands, 07 marking and winged gear chest emblem. Use the reference for his design and proportions. Photograph him as a real, heavy, working metal robot in a real street. He is OUTDOORS on the public pavement beside the curb, in front of a run-down laundromat at night. Rough asphalt road, a gutter, wet cracked concrete, cigarette ends and old brickwork are visible around him. The laundry machines are only seen behind the exterior window glass. He sits on the pavement in his two-leg form, with exactly two arms and two legs, holding a tray of fries. A tired ordinary moment filmed by a friend.
Lighting: Weak greenish fluorescent light leaks through the shop window and mixes badly with a sodium streetlamp. Shop highlights are clipped, metal reflections are irregular and the shadows are muddy.
Texture: Pitted aluminium, scuffed steel joints, flaky orange paint, ingrained oily dirt in the seams, rust around screws, rubbed edges and dried rain marks. His weight creates believable contact shadows on the rough pavement. The fries are ordinary uneven greasy takeaway fries in a stained red-and-white paper tray. Heavy blocky MPEG compression, jagged edges, smeared green-brown colours, chroma noise and interlaced movement artefacts across the entire recording. Fine detail is lost even on the face. The physical scene underneath is photorealistic.

Copy promptGenerate on HiggsfieldCopy & open Higgsfield

@bruiser taking a selfie, photorealistic smartphone photo, shot from the robot's own outstretched arm — his right mechanical forearm enters the bottom-left foreground, huge and slightly out of focus, rust-orange armor plate and worn grey metal filling the corner of the frame.

High-angle selfie POV looking down at @bruiser, front-facing phone camera, ultra-wide 14mm lens with strong barrel/fisheye distortion at the edges, the dome head closest to the lens and exaggerated in size, body tapering down toward the floor.
@bruiser: weathered off-white grey dome helmet with panel seams and rivets, black visor screen with glowing white dot-matrix LED eyes shown as two round dotted circles (happy/curious expression), two antennas with glowing white tips, rust-orange riveted chest armor with stenciled "07" and winged gear emblem, scratches, chipped paint, grime, rust streaks. In his left hand he holds a red-and-white paper carton of golden french fries, held up toward the camera.
Location: empty laundromat late at night, rows of teal front-loading washing machines on both sides, dirty tiled floor with wet streaks, drop ceiling with harsh flickering fluorescent tubes, dark storefront windows in the background with blurry orange street lights.

Lighting: cold overhead fluorescent light, cyan-green color cast, soft glow from the LED visor on the helmet edges, deep shadows under the body.

Camera & texture: real phone camera image, smudged lens with fingerprint marks and haze catching the light, slight motion blur on the foreground arm, sensor noise in the shadows, smartphone HDR processing, mild chromatic aberration at edges, 9:16 vertical framing, candid, amateur, authentic snapshot, ultra-detailed metal textures, physically accurate reflections on the worn metal.

Can you take the camera style based on this prompt that gave me this image and turn it into a reusable MD file?

Copy promptGenerate on HiggsfieldCopy & open Higgsfield

【Generation Goal】
Generate a photorealistic spontaneous office-party recording that clearly demonstrates handheld reaction through the camera's visible behavior. Present a clean full-frame recording image. No camcorder overlay, REC indicator, timestamp, battery icon, focus brackets, viewfinder grid, border or subtitles. The recording device stays outside its own field of view.

【Reference Asset Roles】

@Image 1 supplies only Dan's identity and exact orange quilted vest, green short-sleeved shirt over white long sleeves, beige trousers and off-white sneakers. The sheet's multiple views all show this one man, not additional characters. Copy none of its studio layout or background. Only Dan adopts its facial features.

@Image 2 supplies only the single Bruiser robot's chunky orange-and-silver body, cyan visor eyes, two antennae and physical metal limbs. Copy none of its studio background.

@Audio 1 supplies Dan's vocal identity, tone, timbre, accent and cadence. Use it for Dan's audible English speech in this scene, not as music or playback of the recording's old words.

【Subjects and Relationships】

On-screen cast: exactly one Dan, one Bruiser and one gorilla at the punch bowl. Dan is the only human visible. A separate party guest operates the camera and stays entirely unseen. Dan is in front of the lens, never the person holding the camera. His face, orange vest, both arms and cup are visible throughout.
Keep Dan's orange-vest outfit exclusive to Dan. Each individual keeps their own face and outfit throughout; never use Dan's reference to populate the supporting cast.

【Event Script】
In a wrecked office kitchen, Dan stands on the left and Bruiser stands on the right, both framed from the waist up. Dan holds a tiny paper cup at his own chest, well away from the camera. Bruiser pours champagne from a garden watering can into Dan's cup. It overflows onto Dan's hand and he lifts the wet cup with a startled laugh. Bruiser lowers the can with a sheepish head tilt. A gorilla tends a punch bowl in the background. Finish on Dan's visible amused face and wet cup beside Bruiser.

【Camera and Sound】

Handheld reaction: a third-person medium two-shot, filmed by a separate unseen guest standing in front of Dan and Bruiser. Keep Dan's face and upper body on the left of the picture, Bruiser on the right, and the cup between them. The image has continuous small irregular hand tremor. When Dan's cup suddenly overflows, the camera operator instinctively jerks backward and rolls the frame slightly sideways, then manually corrects the tilted framing to center Dan and Bruiser again. Keep Dan visible while the camera flinches. The operator's hands and cup never enter the foreground. The perspective moves with a person holding the camera; the shot never becomes smooth or stabilized.

Dan audibly speaks a brief surprised English reaction to Bruiser overfilling the cup, then laughs. His visible mouth movements match the actual spoken words. Use @Audio 1 for his vocal identity. Require actual intelligible spoken words; laughter accompanies his speech instead of replacing it. Dan is the only featured speaker. Keep his voice clearly audible above the party noise and synchronize every visible speaking mouth. Preserve natural room noise and the action's own sounds.

One continuous photorealistic shot with natural skin, real clothing and physical props in the same light. An intimate spontaneous recording by an unseen party guest, unpolished consumer-camera footage with visible fine grain. Camera imperfections come from the specified movement, focus or exposure, not graphics on the picture.

【Maintain Consistency】

Preserve the stated cast count, separate identities, outfits, prop ownership, room geography and the action direction throughout. Start as the main action begins, follow the camera behavior once, and finish on its clear result. Present a clean full-frame recording image. No camcorder overlay, REC indicator, timestamp, battery icon, focus brackets,

Copy promptGenerate on HiggsfieldCopy & open Higgsfield

【Generation Goal】
Generate a photorealistic spontaneous office-party recording that clearly demonstrates optical crash zoom through the camera's visible behavior. Present a clean full-frame recording image. No camcorder overlay, REC indicator, timestamp, battery icon, focus brackets, viewfinder grid, border or subtitles. The recording device stays outside its own field of view.

【Reference Asset Roles】

@Image 1 supplies only Dan's identity and exact orange quilted vest, green short-sleeved shirt over white long sleeves, beige trousers and off-white sneakers. The sheet's multiple views all show this one man, not additional characters. Copy none of its studio layout or background. Only Dan adopts its facial features.

@Image 2 supplies only the single Bruiser robot's chunky orange-and-silver body, cyan visor eyes, two antennae and physical metal limbs. Copy none of its studio background.

@Audio 1 supplies Dan's vocal identity, tone, timbre, accent and cadence. Use it for Dan's audible English speech in this scene, not as music or playback of the recording's old words.

【Subjects and Relationships】

On-screen cast: exactly one Dan, one Bruiser and two supporting women: an elderly woman with cropped silver hair in a purple bathrobe and a middle-aged woman with a straight black bob in a cobalt-blue suit. Both women have clearly different facial structures from Dan. No additional party crowd. The operator remains unseen.

Keep Dan's orange-vest outfit exclusive to Dan. Each individual keeps their own face and outfit throughout; never use Dan's reference to populate the supporting cast.

【Event Script】

Bruiser opens a corporate briefcase beside Dan across a boardroom buffet table. Spaghetti spills onto the table as Dan reacts. The two supporting women dance behind them. End with Dan, Bruiser and the open case visible together.

【Camera and Sound】
Optical crash zoom: start with a genuinely wide view of the boardroom showing Dan and Bruiser small across the room. Keep the operator's feet planted in this one location. As the first spaghetti emerges, rapidly change the lens focal length to snap into a medium two-shot of Dan and Bruiser with the open case between them. The background magnifies with the zoom, without forward camera travel. Show a short radial smear, slight framing overshoot and obvious temporary loss of focus, then regain focus with both complete heads and the spilling case in frame. Finish on this settled view.

Dan audibly speaks a brief incredulous English reaction to the spaghetti in the briefcase, then laughs, with synchronized mouth movements. Use @Audio 1 for his vocal identity. Require actual intelligible spoken words; laughter accompanies his speech instead of replacing it. Dan is the only featured speaker. Keep his voice clearly audible above the party noise and synchronize every visible speaking mouth. Preserve natural room noise and the action's own sounds.

One continuous photorealistic shot with natural skin, real clothing and physical props in the same light. An intimate spontaneous recording by an unseen party guest, unpolished consumer-camera footage with visible fine grain. Camera imperfections come from the specified movement, focus or exposure, not graphics on the picture.

【Maintain Consistency】

Preserve the stated cast count, separate identities, outfits, prop ownership, room geography and the action direction throughout. Start as the main action begins, follow the camera behavior once, and finish on its clear result. Present a clean full-frame recording image. No camcorder overlay, REC indicator, timestamp, battery icon, focus brackets,

Copy promptGenerate on HiggsfieldCopy & open Higgsfield

【Generation Goal】

Generate a photorealistic spontaneous office-party recording that clearly demonstrates hidden observation through the camera's visible behavior. Present a clean full-frame recording image. No camcorder overlay, REC indicator, timestamp, battery icon, focus brackets, viewfinder grid, border or subtitles. The recording device stays outside its own field of view.
【Reference Asset Roles】

@Image 1 supplies only Dan's identity and exact orange quilted vest, green short-sleeved shirt over white long sleeves, beige trousers and off-white sneakers. The sheet's multiple views all show this one man, not additional characters. Copy none of its studio layout or background. Only Dan adopts its facial features.

@Audio 1 supplies Dan's vocal identity, tone, timbre, accent and cadence. Use it for Dan's audible English speech in this scene, not as music or playback of the recording's old words.

【Subjects and Relationships】

On-screen cast: exactly one Dan and one tuxedo-wearing raccoon. Dan is the only human; the party doorway and storage-door gap contain no additional people. The operator remains unseen behind the storage-room door.

Keep Dan's orange-vest outfit exclusive to Dan. Each individual keeps their own face and outfit throughout; never use Dan's reference to populate the supporting cast.

【Event Script】

In a copier alcove beside the office party, Dan slips cash under a table to a tuxedo-wearing raccoon, who slides him a tiny gold trophy. Dan glances toward the separate party doorway away from the hidden operator while taking the trophy. The observer peeks and retreats behind the storage-room door. Dan and the raccoon remain unaware of the person filming.

【Camera and Sound】

Hidden observation: the operator stays concealed behind a partly closed storage-room door, filming through a narrow gap toward the copier alcove. Dark door edges cover much of the foreground. Slowly lean sideways just enough to uncover Dan, the raccoon and the cash exchange, then pull back behind the door again. The camera peeks from cover rather than tracking openly. Dan glances toward a different party doorway on the opposite side of the alcove, away from the concealed lens, remaining unaware of the filming.

Dan quietly speaks a brief conspiratorial English reaction to the raccoon during their exchange. His speech is audible without addressing or noticing the hidden operator. Use
@Audio 1 for his vocal identity. Require actual intelligible spoken words; laughter accompanies his speech instead of replacing it. Dan is the only featured speaker. Keep his voice clearly audible above the party noise and synchronize every visible speaking mouth.
Preserve natural room noise and the action's own sounds.

One continuous photorealistic shot with natural skin, real clothing and physical props in the same light. An intimate spontaneous recording by an unseen party guest, unpolished consumer-camera footage with visible fine grain. Camera imperfections come from the specified movement, focus or exposure, not graphics on the picture.

【Maintain Consistency】

Preserve the stated cast count, separate identities, outfits, prop ownership, room geography and the action direction throughout. Start as the main action begins, follow the camera behavior once, and finish on its clear result. Present a clean full-frame recording image. No camcorder overlay, REC indicator, timestamp, battery icon, focus brackets,

Copy promptGenerate on HiggsfieldCopy & open Higgsfield

【Generation Goal】
Generate a photorealistic spontaneous office-party recording that clearly demonstrates reactive whip pan through the camera's visible behavior. Present a clean full-frame recording image. No camcorder overlay, REC indicator, timestamp, battery icon, focus brackets, viewfinder grid, border or subtitles. The recording device stays outside its own field of view.

【Reference Asset Roles】
@Image 1 supplies only Dan's identity and exact orange quilted vest, green short-sleeved shirt over white long sleeves, beige trousers and off-white sneakers. The sheet's multiple views all show this one man, not additional characters. Copy none of its studio layout or background. Only Dan adopts its facial features.

@Audio 1 supplies Dan's vocal identity, tone, timbre, accent and cadence. Use it for Dan's audible English speech in this scene, not as music or playback of the recording's old words.

【Subjects and Relationships】

On-screen cast: exactly one Dan, an older woman with long blonde hair, exposed shoulders and bare arms wearing ONLY a sleeveless white wedding dress and an elderly bald man with a full grey beard in a navy suit. The woman's entire torso is white bridal satin, with no orange vest, no green shirt and no undershirt. Her face has a broad jaw and deep smile lines, unlike Dan. The woman rides a small actual motorcycle with two motorcycle wheels and handlebars; the bearded man is seated in its sidecar. The three people have different faces and hairstyles. Streamers and party debris suggest the gathering without additional people. The operator remains unseen.

Keep Dan's orange-vest outfit exclusive to Dan. Each individual keeps their own face and outfit throughout. Dan alone has any orange or green garment. The bride keeps bare shoulders and bare arms throughout, with the sleeveless white gown as her only clothing. The sidecar man's entire outfit is navy. Never use Dan's reference to dress either rider or supply their faces.

【Event Script】

Dan stands beside an improvised office bar. An engine rev makes him turn toward the sound. The woman in a wedding dress rides the small motorcycle with the navy-suited bearded man in its sidecar past the operator and along the office aisle. The camera turns to catch them passing and settles as they continue away.

【Camera and Sound】
Reactive whip pan: start in a medium close view of Dan at the office bar. When the motorcycle engine revs, respond with one rapid horizontal rotation toward the aisle. Intervening desks smear sideways across the frame during the turn. Overshoot the passing riders slightly, correct once, then settle on the wedding-dressed rider and navy-suited sidecar passenger continuing down the aisle. This is a physical camera turn in the same continuous space, not a cut, zoom or forward push. Dan's spontaneous reaction remains audible after he leaves the frame.

Dan audibly speaks a brief surprised English reaction as the engine interrupts him. Continue hearing his voice naturally from beside the operator when the camera turns away from his face. Use @Audio 1 for his vocal identity. Require actual intelligible spoken words; laughter accompanies his speech instead of replacing it. Dan is the only featured speaker. Keep his voice clearly audible above the party noise and synchronize every visible speaking mouth. Preserve natural room noise and the action's own sounds.

One continuous photorealistic shot with natural skin, real clothing and physical props in the same light. An intimate spontaneous recording by an unseen party guest, unpolished consumer-camera footage with visible fine grain. Camera imperfections come from the specified movement, focus or exposure, not graphics on the picture.

【Maintain Consistency】

Preserve the stated cast count, separate identities, outfits, prop ownership, room geography and the action direction throughout. Start as the main action begins, follow the camera behavior once, and finish on its clear result. Present a clean full-frame recording image. No camcorder overlay, REC indicator, timestamp, battery icon, focus brackets,

Copy promptGenerate on HiggsfieldCopy & open Higgsfield

【Generation Goal】
Generate a photorealistic spontaneous office-party recording that clearly demonstrates walking backward through the camera's visible behavior. Present a clean full-frame recording image. No camcorder overlay, REC indicator, timestamp, battery icon, focus brackets, viewfinder grid, border or subtitles. The recording device stays outside its own field of view.
【Reference Asset Roles】
@image1 supplies only Dan's identity and exact orange quilted vest, green short-sleeved shirt over white long sleeves, beige trousers and off-white sneakers. The sheet's multiple views all show this one man, not additional characters. Copy none of its studio layout or background. Only Dan adopts its facial features.
@image2 supplies only the single Bruiser robot's chunky orange-and-silver body, cyan visor eyes, two antennae and physical metal limbs. Copy none of its studio background.
@audio1 supplies Dan's vocal identity, tone, timbre, accent and cadence. Use it for Dan's audible English speech in this scene, not as music or playback of the recording's old words.

【Subjects and Relationships】
On-screen cast: exactly one Dan and one Bruiser. Dan is the only human in the aisle. The operator remains unseen.
Keep Dan's orange-vest outfit exclusive to Dan. Each individual keeps their own face and outfit throughout; never use Dan's reference to populate the supporting cast.

【Event Script】
Dan advances through an office aisle carrying an enormous gold trophy full of punch. Bruiser follows just behind his shoulder and makes one clumsy attempt to scoop punch with an oversized ladle. Dan keeps walking toward the retreating observer and laughs. Finish still facing Dan, with Bruiser behind him and the punch trophy clearly visible.

【Camera and Sound】
Walking backward: Dan walks directly toward the lens while the operator physically retreats down the same office aisle. Keep Dan's face, punch trophy and Bruiser behind his shoulder in view. Each backward footstep produces uneven vertical bob and a small roll; nearby cubicle edges slide past faster than the distant background. Correct the framing once while continuing backward. Dan's size stays approximately steady because both people move. Do not replace this retreat with a stationary lens zoom or a smooth tracking rig.
Dan audibly speaks a brief amused English reaction to Bruiser trying to scoop his punch while he walks, with synchronized mouth movements. Use @audio1 for his vocal identity. Require actual intelligible spoken words; laughter accompanies his speech instead of replacing it. Dan is the only featured speaker. Keep his voice clearly audible above the party noise and synchronize every visible speaking mouth. Preserve natural room noise and the action's own sounds.
One continuous photorealistic shot with natural skin, real clothing and physical props in the same light. An intimate spontaneous recording by an unseen party guest, unpolished consumer-camera footage with visible fine grain. Camera imperfections come from the specified movement, focus or exposure, not graphics on the picture.

【Maintain Consistency】
Preserve the stated cast count, separate identities, outfits, prop ownership, room geography and the action direction throughout. Start as the main action begins, follow the camera behavior once, and finish on its clear result. Present a clean full-frame recording image. No camcorder overlay, REC indicator, timestamp, battery icon, focus brackets,

Copy promptGenerate on HiggsfieldCopy & open Higgsfield

【Generation Goal】
Generate a photorealistic spontaneous office-party recording that clearly demonstrates foreground obstruction and autofocus recovery through the camera's visible behavior. Present a clean full-frame recording image. No camcorder overlay, REC indicator, timestamp, battery icon, focus brackets, viewfinder grid, border or subtitles. The recording device stays outside its own field of view.

【Reference Asset Roles】
@image1 supplies only Dan's identity and exact orange quilted vest, green short-sleeved shirt over white long sleeves, beige trousers and off-white sneakers. The sheet's multiple views all show this one man, not additional characters. Copy none of its studio layout or background. Only Dan adopts its facial features.
@audio1 supplies Dan's vocal identity, tone, timbre, accent and cadence. Use it for Dan's audible English speech in this scene, not as music or playback of the recording's old words.

【Subjects and Relationships】
On-screen cast: exactly one Dan, one tuxedo-wearing raccoon and an older woman with short silver curls wearing the purple blazer and carrying the blue folders. Her face is clearly unlike Dan's. The operator remains unseen.
Keep Dan's orange-vest outfit exclusive to Dan. Each individual keeps their own face and outfit throughout; never use Dan's reference to populate the supporting cast.

【Event Script】
Dan plays beer pong against a tuxedo-wearing raccoon on a messy office table. He misses a throw as a coworker in a purple blazer crosses close to the lens carrying bright blue folders. The folders briefly hide Dan, then clear. Dan laughs and gestures at the raccoon. The coworker continues out of view.

【Camera and Sound】
Brief foreground obstruction and focus recovery: start looking across the beer-pong table at Dan and the raccoon. The purple-clad coworker makes one continuous crossing directly in front of the lens. Bright blue folders fill the near foreground and briefly block Dan. Autofocus briefly grabs the close folders, leaving Dan blurry. After her folders have cleared the lens completely, the operator immediately corrects focus directly onto Dan's eyes and speaking mouth. Dan's face visibly changes from blurry to crisp, with his moustache, eyes and clothing seams clearly resolved. Hold Dan sharply in the final composition; the raccoon stays slightly behind him. The operator reframes slightly upward to center Dan's face during recovery. Keep the same viewpoint and uninterrupted moment.
Dan audibly speaks a brief amused English reaction to his missed throw and the raccoon as the folders clear, with synchronized mouth movements. Use @audio1 for his vocal identity. Require actual intelligible spoken words; laughter accompanies his speech instead of replacing it. Dan is the only featured speaker. Keep his voice clearly audible above the party noise and synchronize every visible speaking mouth. Preserve natural room noise and the action's own sounds.
One continuous photorealistic shot with natural skin, real clothing and physical props in the same light. An intimate spontaneous recording by an unseen party guest, unpolished consumer-camera footage with visible fine grain. Camera imperfections come from the specified movement, focus or exposure, not graphics on the picture.

【Maintain Consistency】
Preserve the stated cast count, separate identities, outfits, prop ownership, room geography and the action direction throughout. Start as the main action begins, follow the camera behavior once, and finish on its clear result. Present a clean full-frame recording image. No camcorder overlay, REC indicator, timestamp, battery icon, focus brackets,

Copy promptGenerate on HiggsfieldCopy & open Higgsfield

【Generation Goal】
A photorealistic candid recording at an absurd office party, demonstrating the camera adjusting from a dark room to harsh rooftop sunlight.

【Reference Asset Roles】
@image1 is only Dan's character sheet: keep his face and orange quilted vest, green short sleeves over white long sleeves, beige trousers and off-white shoes. The views all show the same single Dan. Create a new scene.
@audio1 supplies Dan's voice, accent and cadence for his audible English reaction, not the old recording's words.
【Subjects and Relationships】
Four different people: Dan; his elderly bald boss with a round face, full grey beard and purple bathrobe; a tall black-curled woman in a blue suit; a short silver-haired older woman in a burgundy jumpsuit. Only Dan resembles @image1 or wears his outfit. The boss sits in a blue inflatable pool of cereal; the women stand beside a shopping cart carrying a disco ball.

【Event Script】
Dan carries a tray of paper cups out of a dim office storeroom onto the sunlit rooftop. His boss is directly ahead. Dan spots him in the cereal pool, looks down at the boss's face, and gives him an incredulous amused spoken reaction in English. He keeps walking toward the boss and laughs while still looking at him.

【Camera and Sound】
One uninterrupted handheld shot from behind Dan's left shoulder, following him through the doorway. Keep the boss in front of Dan, along his gaze. Outdoors we see the back and left side of Dan's head; his head, eyes and shoulders stay directed at the boss while he speaks. He never turns toward the operator.
The camera starts correctly exposed for the dark room. Entering sunlight briefly washes the entire image nearly white. Automatic exposure then visibly lowers the brightness until Dan and the cereal pool become readable together. Natural footstep bob, small framing corrections and fine film grain. Audible Dan speech using @audio1, followed by laughter and rooftop ambience.

【Maintain Consistency】
Keep all four identities and outfits distinct, the cups in Dan's hands and his gaze on the boss throughout the outdoor reaction. Clean camera image without REC, timestamps, viewfinder graphics, text or subtitles. The filming camera and operator stay out of view.

Copy promptGenerate on HiggsfieldCopy & open Higgsfield

【Generation Goal】
Generate a photorealistic spontaneous office-party recording that clearly demonstrates a mockumentary-style awkward zoom through the camera's visible behavior. Present a clean full-frame recording image. No camcorder overlay, REC indicator, timestamp, battery icon, focus brackets, viewfinder grid, border or subtitles. The recording device stays outside its own field of view.

【Reference Asset Roles】
<<<image_1>>> supplies only Dan's identity and exact orange quilted vest, green short-sleeved shirt over white long sleeves, beige trousers and off-white sneakers. The sheet's multiple views all show this one man, not additional characters. Copy none of its studio layout or background. Only Dan adopts its facial features.
<<<image_2>>> supplies only the single Bruiser robot's chunky orange-and-silver body, cyan visor eyes, two antennae and physical metal limbs. Copy none of its studio background.
<<<audio_1>>> supplies Dan's vocal identity, tone, timbre, accent and cadence. Use it only for Dan's audible English speech in this scene, not as music, not for Bruiser, and not as playback of the recording's old words.

【Subjects and Relationships】
On-screen cast: exactly one Dan, one Bruiser and one gorilla at the punch bowl. Dan is the only human visible. A separate party guest operates the camera and stays entirely unseen. Dan is in front of the lens, never the person holding the camera. His face and orange vest are visible throughout.
Keep Dan's orange-vest outfit exclusive to Dan. Each individual keeps their own face and outfit throughout; never use Dan's reference to populate the supporting cast. The cable bouquet belongs to Bruiser at the start and to Dan after the handover.

【Event Script】
In the same wrecked office kitchen, Dan stands on the left and Bruiser stands on the right, both framed from the waist up. A gorilla tends a punch bowl in the background, ladling punch throughout. Bruiser holds out a bouquet made of tangled colorful electrical cables, their plugs and connectors fanned out like flower heads, the stems bound with silver tape. Bruiser says "Here is a gift." Dan takes the bouquet with both hands and says "Oh, thank you, Bruiser." Dan pauses, holding the cable bouquet at his chest, then slowly turns his head and looks directly into the lens with an awkward, uncomfortable half-smile. Bruiser stays beside him with a proud head tilt. Finish on Dan's awkward face staring into the lens, holding the bouquet.

【Camera and Sound】
Mockumentary handheld: a third-person medium two-shot, filmed by a separate unseen guest standing in front of Dan and Bruiser. Keep Dan on the left of the picture, Bruiser on the right, and the bouquet between them during the handover. The image has continuous small irregular hand tremor. The moment Dan looks into the lens, the camera performs a fast, slightly clumsy optical zoom toward Dan's face, ending in a tight close-up. The camera body stays where it is; only the lens zooms in. During the zoom the autofocus fails: Dan's face goes soft and blurry, the focus hunts past him once, then snaps back sharp on his awkward expression. Bruiser drifts out of frame on the right as the zoom tightens. The operator's hands never enter the foreground. The shot never becomes smooth or stabilized.
Bruiser speaks first with a distinct synthetic robotic voice, clearly different from Dan's: "Here is a gift." Dan then answers in English: "Oh, thank you, Bruiser." Use <<<audio_1>>> for Dan's vocal identity only. Require actual intelligible spoken words for both lines. Each line comes only from its own speaker, and Bruiser's visor flickers in time with his speech. Dan's visible mouth movements match his spoken words exactly. During the awkward look and zoom, there is no dialogue. Keep both voices clearly audible above the party noise.
Preserve natural room noise, a faint lens-zoom motor whir and the action's own sounds.
One continuous photorealistic shot with natural skin, real clothing and physical props in the same light. An intimate spontaneous recording by an unseen party guest, unpolished consumer-camera footage with visible fine grain. Camera imperfections come from the specified movement, zoom and focus failure, not graphics on the picture.

【Maintain Consistency】
Preserve the stated cast count, separate identities, outfits, prop ownership, room geography and the action direction throughout. Start as Bruiser offers the bouquet, follow the zoom and focus failure once, and finish on Dan's awkward look into the lens. Present a clean full-frame recording image. No camcorder overlay, REC indicator, timestamp, battery icon, focus brackets, viewfinder grid, border or subtitles. The recording device stays outside its own field of view.