Explore Skills

humanize-ai-text

Emma

This prompt instructs the assistant to act as a skilled human editor who rewrites a provided text in a natural, credible style while preserving all original meaning and factual details. It specifies a formal or informal tone, improves clarity and flow, removes robotic or repetitive phrasing, and enforces constraints such as not inventing information, keeping names and numbers intact, and using clean paragraph-based formatting. The result is a polished rewrite that reads smoothly to the intended audience and avoids common AI writing signals.

This prompt instructs an image-editing workflow to transform a provided photo into a square, photoreal instant-film style portrait with a Polaroid-like frame, replacing the background with a softly wrinkled white fabric curtain and adjusting the subjects’ bodies into a very tight hug while keeping both faces exactly the same as the supplied face references. It applies a dim-room direct-flash look with slight motion blur, soft contrast, light vignette, and moderate grain, then reinforces and verifies facial identity with multiple identity-lock modules and automatic retries, and if identity similarity still falls short it composites the original faces back in as a failsafe, producing a clean output with no added text, logos, watermarks, or new props.

female-only-beauty-editorial-image-preview-screen

Emma

This prompt directs an AI image workflow to create or refine a close-up view of a MacBook Pro screen showing two side-by-side beauty portraits in macOS Preview, emphasizing a realistic behind-the-scenes image review moment. It specifies consistent high-end editorial makeup, lighting, background, and accessories across both portraits while preserving the subject’s exact facial identity and natural skin texture, with only subtle pose differences between the two images. The intended outcome is a clean, professional, readable screen display where both portraits remain clearly comparable, with realistic UI details and reflections and without stylized effects or unnecessary distractions.

victorias-secret-backstage-angel

Emma

This prompt instructs an image-generation/editing model to create an ultra-realistic, high-fashion backstage portrait that preserves the subject’s exact face, hairstyle, and expression from an input photo while placing them in a glamorous runway-prep environment. It specifies a waist-up, diagonally angled composition with a makeup artist’s hands applying lipstick, plus detailed wardrobe elements like a crystal-embroidered corset and feathered wings, all captured with realistic on-camera flash, cinematic backstage ambience, and crisp texture fidelity. The intended outcome is an 8K, 4:5 editorial-style image that looks like an authentic pre-runway moment—luxurious, sensual, and confidently poised—without altering the subject’s facial identity.

This prompt instructs an image-to-image model to transform a supplied portrait into a polished, LinkedIn-ready headshot that strongly preserves the subject’s identity. It specifies professional studio-style composition, lens and depth-of-field characteristics, Rembrandt-inspired lighting, a seamless neutral background with subtle gradient, restrained corporate color grading, and texture-preserving retouching while explicitly avoiding common artifacts (distortion, oversmoothing, busy backgrounds, and harsh effects). The intended result is a clean, credible, approachable head-and-shoulders headshot delivered in platform-friendly square and vertical output sizes.

This prompt directs an AI image generator to produce a photorealistic scene featuring a commercial 1/7 scale figurine based on the characters shown in the provided image, displayed on a computer desk with a clear round acrylic base. It specifies environmental details and supporting props, including a computer monitor showing the figurine being modeled in 3D and a nearby premium-style collectible packaging box decorated with original 2D illustration artwork, resulting in a realistic product-style tabletop display.

surveillance-style-cctv-portrait

Emma

This prompt instructs an image model to generate a vertical 4:5, high-mounted retail CCTV still showing a person behind a cluttered checkout counter, shot from a steep top-down angle with slight fisheye distortion. It specifies a defiant pose with an upward gaze and a raised hand, while keeping facial identity unreadable under oversized dark sunglasses and an open mouth, all under harsh fluorescent lighting. The intended outcome is a convincingly unedited, low-fidelity surveillance capture with heavy motion blur, compression artifacts, noise, scan lines, banding, and reduced dynamic range, set in a cramped, chaotic shop interior without any cinematic polish or beautification.

winter-solitude-triptych-cinematic-portrait

Emma

This prompt instructs an image-generation system to produce a 2160×3840 vertical 4K photorealistic triptych made of three equal horizontal panels that tell a cold, cinematic story of melancholy in a snowy landscape. It specifies consistent lighting, color grading, camera/lens characteristics, wardrobe, snowfall, and a panel-by-panel composition (portrait with transparent umbrella, isolated full-body walk, and an emotional eye close-up), while requiring the subject’s identity to match an accompanying reference image exactly without any facial or anatomical changes. The expected outcome is a cohesive, film-grain–treated fine-art triptych that preserves true likeness and delivers an introspective narrative of loneliness and nostalgia across the three frames.

notre-dame-saigon-night

Shaun The Sheep

This prompt instructs an image-editing model to take the user’s provided portrait photo and composite it into a new photorealistic nighttime scene that matches the referenced Notre-Dame Cathedral (Ho Chi Minh City) location for background, lighting, and atmosphere, while keeping the user’s facial identity exactly unchanged. The expected result is a realistic night portrait of the same person standing on a balcony with a black-painted railing in a specified side-profile pose, featuring warm golden illumination, accurate shadows/reflections, and correct perspective consistent with the reference environment.

ai-text-detection

Emma

This prompt instructs an AI to evaluate a provided text for signs of AI, human, or mixed authorship by examining observable linguistic, structural, semantic, and repetition-based features, then assign a composite AI-likelihood score with evidence-based indicators, a confidence level, and a final classification in a fixed, interpretable output format.

behind-the-scenes-studio-portrait

Emma

This prompt instructs an image generator to create a realistic behind-the-scenes photograph of a professional studio during a tethered portrait session, emphasizing the technical workflow and equipment. It specifies a foreground monitor and laptop that both display the same high-resolution, softly lit portrait of the subject (with the subject’s face and hair preserved exactly from a provided reference image), along with believable studio lights, reflectors, cables, and a subtle background assistant. The intended result is a clean, professional, documentary-style scene that contrasts the polished portrait on-screen with the practical working environment, without stylization, clutter, or any alteration of the subject’s identity.

casual-daytime-car-selfie

Emma

This prompt directs an image generator to produce a candid, phone-style, chest-up selfie of a casually styled person sitting in the driver’s seat of a modern car with the sunroof open in bright natural daylight, sipping a clear matcha latte through a straw while looking at the camera. It specifies wardrobe, grooming, accessories, framing, lighting, and background details (palm trees, blue sky, soft urban greenery blur) to ensure a realistic, unposed lifestyle moment, and it explicitly avoids studio polish, heavy retouching, full-body framing, or artificial-looking elements.

This prompt configures an image-generation workflow to produce a high-resolution black-and-white studio portrait with cinematic lighting and minimal styling while strictly preserving the subject’s original identity, gender presentation, and exact hairstyle from a provided reference (including required face/hair mattes and guides), applying only controlled monochrome tonal grading; if required reference materials are missing or any identity/gender/hair drift is detected against specified similarity metrics, the process aborts rather than outputting an altered result.

Prompt này cấu hình một quy trình tạo ảnh mirror selfie chân thực từ ảnh tham chiếu, trong đó khuôn mặt người và chiếc iPhone (đặc biệt là màu cam và cụm camera) được “khóa” theo ảnh mẫu với mức sai khác cực thấp, áp dụng denoise theo vùng để giữ chi tiết quan trọng, thiết lập bố cục/ánh sáng tự nhiên như chụp bằng smartphone, thêm các ràng buộc và tiêu chí kiểm định để tự động loại kết quả lệch mặt hoặc sai thiết kế điện thoại; nếu thiếu ảnh tham chiếu bắt buộc thì dừng và yêu cầu người dùng tải ảnh.

This prompt instructs the AI to produce a structured feature specification document for a proposed product feature, using provided inputs such as target users, key functionalities, technical considerations, and estimated development effort. The resulting output is a clear, implementation-ready spec organized into standard product and engineering sections—overview, goals with success metrics, user stories, acceptance criteria mapped to those stories, a high-level technical design, and future considerations—so developers can begin building with aligned requirements and scope.

stranger-things-group-fisheye-selfie

Emma

This prompt instructs an image editor to use a provided reference photo and perform a precise face-swap on only the front-center person taking the selfie, replacing that person’s face with the user’s face while preserving the user’s exact identity (including hair and natural realism) and matching the original serious expression. All other characters, their poses, and the full 1980s living-room setting, lighting, fisheye selfie perspective, and cinematic styling must remain unchanged, producing a photorealistic composite where only the central selfie-taker’s face is different.

2000s-mirror-selfie-cinematic-nostalgia

Emma

This prompt instructs an image model to generate a hyper-realistic, early-2000s-style mirror selfie with a cinematic, nostalgic mood, emphasizing an unblocked facial reflection, direct on-camera flash, dim minimal background, and tactile analog imperfections like grain, dust, and light scratches as if printed on glossy photo paper. The intended outcome is a vertically framed (3:4) portrait that feels emotionally candid and timeworn—natural hair, calm confident expression, satin-shirt highlights, and subtle mirror/flash artifacts—while avoiding modern polish, branding, clutter, or provocative styling.

leica-monochrome-duality

Huynh Huong

This prompt instructs an AI image generator to produce an ultra-realistic, 8K high-fashion editorial portrait in stark black and white, using a minimalist studio setup with sculptural single-source lighting and a clean light background. It specifies tight framing, Leica-style optics and depth of field, matte monochrome tonality with fine grain, and gender-specific wardrobe/grooming options, while requiring the subject’s identity to match the provided reference exactly without facial reshaping or heavy retouching. The result is a timeless, high-contrast monochrome portrait with razor-sharp eyes, authentic skin and fabric texture, and an introspective, confident mood.

three-generation-family-studio-portrait-prompt

Shaun The Sheep

This prompt instructs an AI to act as a professional studio family portrait director and produce a single polished, photorealistic multi‑generation group portrait by seamlessly merging the provided individual reference photos into one cohesive studio scene. It specifies the desired studio backdrop color, lighting style, and coordinated wardrobe, and requires consistent shadows and color grading, authentic expressions, and preserved facial likeness and proportions for every person. The intended outcome is an elegant, timeless, print‑ready family portrait that includes exactly all original family members with no additions or omissions, avoids unrealistic edits or props, and follows privacy, workflow, and quality constraints to ensure a natural, high‑end result delivered reliably.

winter-dive-goblin-inspired-triptych-portrait

Emma

This prompt instructs an image-generation system to create a high-resolution, photorealistic three-panel (triptych) cinematic winter portrait series of the same person, using an unchanged likeness from a provided reference image. It specifies a serene snowy mountain-and-lake setting with light snowfall and soft overcast lighting, and defines each panel’s viewpoint (rear, profile, and close-up), wardrobe, props, mood, and consistent color/texture treatment so the final output forms a visually harmonized, introspective editorial-style sequence.