Canvex

Use this skill when the user asks for a multi-image Amazon listing pack — typical surface forms include "做一套 Amazon 主图", "7 张套图", "Amazon listing 套图", "上架图包", "Amazon 7-pack", "listing image set". Generates the canonical 7-image set in PARALLEL — main shot, infographic, alternate angle, detail close-up, two lifestyle scenes, and scale reference — all sharing ONE img2img source so the same product carries across every angle. Open-ended prompts let img2img decide product-specific details (cup → handle close-up, wall art → back hook, rug → underside texture).

Use this skill whenever the user asks to generate, create, draw, or visualize an image on the canvas — including product photos, lifestyle scenes, mockups, marketing visuals, concept art, illustrations, or icons. The skill rewrites vague requests into high-quality prompts and picks the right size + count before calling `generate_image`.