What problem does it solve? Creating professional product photography for e-commerce, social media, and paid ads normally requires studio shoots or skilled prompt engineering. This Skill routes product image requests through the Higgsfield product-photoshoot CLI, whose backend prompt enhancer applies mode-specific photography vocabulary and structural templates before submitting to gpt_image_2, producing consistent brand-quality visuals without manual prompt writing. ## Core Features & Use Cases - Ten generation modes: product_shot, lifestyle_scene, closeup_product_with_person, moodboard_pin, hero_banner, social_carousel, ad_creative_pack, virtual_model_tryout, conceptual_product, and restyle, selected by user intent with documented tie-breakers. - Structured pre-generation interview: Asks at most four labeled-option questions based on request type (uploaded photo, named use case, text-only, restyle, model tryout, vague request) before submitting. - Multi-variant and carousel support: The --count flag returns distinct variants, and carousel or ad-pack modes lock a consistent visual system across all slides. - Use Case: A seller uploads a candle photo and asks for Pinterest content; the Skill selects moodboard_pin mode, runs higgsfield product-photoshoot create with the image, and returns the generated image URLs. ## Quick Start Ask the assistant to create three lifestyle product shots from your uploaded product photo using the higgsfield product-photoshoot command.
