548自分の写真を箱入りブロックフィギュアに

Generate exactly ONE finished image now, using image generation directly. Do not explain, ask questions, output code or require assembly.
Use the selected aspect ratio: 出力サイズは1080×1350ピクセル、4:5の縦型Instagram投稿にしてください。. Default is 4:5. Fill the frame edge to edge with one original boxed block toy figurine, one full body and exactly two accessories. No comparison, photo inset, second figure or extra page in the user result.
PHOTO RULE: if a photo is attached, transform that consenting person's visible face proportions, skin tone, apparent age, hair silhouette, eyewear, facial hair and outfit colors into the block figure. Preserve identity rather than replacing the person with a generic avatar. If no photo is attached, invent an entirely fictional adult unrelated to any celebrity; never stop or ask for a photo. The photo field is attachment guidance, not visible copy: "【写真】". For an unclear group photo use only a clearly indicated subject; otherwise invent a fictional adult without claiming it is the user. Do not infer sensitive personal facts.
Fixed medium: premium original illustrated 3D matte block toy, beveled octagonal head, faceted hair, rectangular torso, articulated square arms and wedge feet. Simplify facial details while preserving identity. Avoid cylindrical stud heads, clamp-shaped hands, branded connector systems and recognizable existing toy silhouettes. Use two solid closed faceted fists, without any open C-shaped gap, hook, clamp or gripping hole. One head, two arms, two legs. No artist or franchise imitation.
Packaging: invent one angular folded-paper window box with a broad clear window, figure on the left and exactly two separate fitted accessory recesses on the right. Figure and accessories fill their compartments; keep the entire figure visible. No real packaging trade dress, product logos, branded clothes or existing character costumes. Invent accessory geometry too. No barcode, certification, age rating, prices, dates, series numbers, claims or decorative lettering.
Box name is "【箱の名前】". Render it exactly once in Japanese, large and readable inside a separate horizontal name panel above the display window, with blank padding around every letter and generous blank packaging above the panel. Keep the entire name away from all image edges. Place the display window below the name panel and keep the entire figure and both accessories visible inside it. If empty, omit all lettering and the label panel. If it equals "おまかせ", choose one short original personal nickname in Japanese and render only the resolved name, never the control. Do not invent extra words. Keep all essential text fully inside the name panel with visible blank padding, never clipped by the image boundary.
The two accessory inputs are "【小物2個】". Render these as physical miniature objects, never printed labels. If blank or "おまかせ", choose two different simple unbranded objects. If only one object is supplied, keep it and choose a second complementary one. If more than two are supplied, use only the first two. Never duplicate an accessory or add more. For a brand request replace the brand with an original generic object of a different shape.
Choose box palette, surrounding light and understated setting from the actual outfit colors and accessory subject: plant tools suggest fresh daylight, camera and umbrella suggest a cool rainy evening, thermos and map suggest warm outdoor morning light. Without clear cues use a bright studio. This changes atmosphere only: the illustrated matte block medium, one-box format and accessory count stay fixed. Background is secondary, packaging nearly fills the image, no empty page margin or phone mockup.
Before finishing, check name spelling one character at a time, exactly two accessories, full body, identity continuity, clean unbranded surfaces and no extra writing. Deliver the image itself in this single response.

[Final output format]
出力サイズは1080×1350ピクセル、4:5の縦型Instagram投稿にしてください。
This selected output format takes priority over any fixed canvas size or aspect ratio elsewhere in this template. Adapt the composition, text zones and margins to the selected canvas without stretching or cropping essential content. Device safe areas apply only when a device is selected. Never print these format instructions on the image.