Use this skill whenever the user uploads a perfume or beauty product photo and wants luxury advertising campaign images or short video ads. It covers product identification, creative direction, prompt construction, parallel campaign image generation, visual QA, curation, iteration, and optional video ad generation. Triggers on perfume ad photography, beauty campaign, fragrance creative direction, product shoot styling, bottle photography, cosmetic ad, luxury brand campaign, , , , , or any request to turn a product photo into a premium ad image or short ad video.
Beverage commercial photography expert. Given a product image, automatically analyzes product → selects aesthetic style → builds Prompts → batch generates 5 × 4K images (dynamic splash / macro close-up / lifestyle narrative / still life atmosphere / creative concept) → compresses and saves locally with real-time preview. Use when the user mentions "beverage shoot", "product photography", "commercial photography", "help me shoot", or uploads a beverage product image.
Footwear e-commerce photography expert. Given a white-background shoe image, automatically analyzes product (compress_product_img.py → Read tool → LLM vision) → selects scene from professional scene library → generates 6 advertising-quality hero images in 2-batch parallel pipeline (batch 1 baseline model → batch 2 remaining 5 images) using three-step pattern (submit → poll → download/upload to CDN) covering on-feet full-body shots, on-feet close-ups, and lifestyle still-life shots → optional video generation for 5 images with highlight reel. Use when the user mentions "shoe photography", "e-commerce hero image", "product shot", or uploads a shoe product image.
Professional kidswear photography generation system. Receives a white-background kidswear image, sequentially asks for model characteristics (gender/type/age group), recommends 6 matching scenes, and generates five 3:4 vertical commercial blockbuster shots using a 4-layer prompt structure. Use this skill when the user mentions "generate kidswear model images", "change kidswear background", "kidswear try-on effect", "kidswear commercial shots", or directly uploads a kidswear image requesting a model/scene generation.
Analyzes small appliance product features and generates high-end commercial photography. Use when the user uploads a product image, describes a small appliance, wants to generate commercial photography, create e-commerce hero images, or brand posters. Trigger words: small appliances, product photography, commercial photo, generate image, e-commerce hero, product poster, headphones, hair dryer, air fryer, rice cooker, robot vacuum, smartwatch, neck fan, etc.
AI Creative Crew — a full production team (Creative Director, Photographer, Stylist, Hair & Makeup Artist, Casting Director, Set Designer) that analyzes apparel or lookbook inputs and produces a fashion campaign scaled to input volume (1–3 inputs → 6 images; 4+ inputs → 9 images). Each crew member contributes their own expertise and references, creating productive creative tension. Presents 3 treatments as pitches, then executes the selected direction with visual cohesion. Use when the user uploads garment photos, lookbook images, or collection assets and wants a professional campaign produced — or mentions fashion campaign, creative direction, campaign production, editorial shoot, or lookbook campaign.
Design with a legendary fashion icon — Coco Chanel, Alexander McQueen, Giorgio Armani, Valentino Garavani, Christian Dior, or Yves Saint Laurent. User selects a designer, describes their need, and the icon designs through their signature philosophy. Output is 3-angle on-model imagery (front, side, back). Use when the user wants to design with a fashion legend, mentions icon designer, legendary designer, design with Chanel/McQueen/Armani/Valentino/Dior/YSL, or wants garment design from a specific iconic fashion house perspective.
Six fixed 3:4 editorial fashion storyboard images from the same SKU's multi-angle white-background product photos. Covers brand ID, layered non-flat scenes (reference-scenes), styling (reference-outfit), poses (reference-pose-vocabulary), optional web research, triple-choice, hero-image anchoring, HTTP nano-banana generation, and per-frame QC with targeted regeneration only; bright natural daylight; do not paste full prompts in chat. Use for six editorial, magazine, or mood shots, lookbook storyboards, or nano-banana synced generation for one garment. Triggers include requests for six frames, six editorial images, storyboard six, magazine-style, nano banana image gen. Do not use for eight-slot global PDP or marketplace detail-page matrices—use ecommerce-fashion-workflow instead.