3D Figurine From Photo

Use the 3D AI Studio MCP to run the "Photo to Figurine" template (https://www.3daistudio.com/templates/3d-figurine-from-photo) for me.

My photo(s): I'll hand them over when you call upload_image.
Subject: Person
Style: Stylized
Base: Classic
Pose: Original

Follow these steps in order. The prompts and settings are the template's own, so use them word for word.

1. Get my photos
Call upload_image with purpose "Photo for the Photo to Figurine template" so I can hand them over (images attached in the chat don't reach the MCP). I can upload up to 5 photos of the same subject; use all of them.

2. Prepare my photo
Call edit_image_gpt with my uploaded photo(s) in images, quality "low", output_format "png" and this prompt:
"""
The reference photo(s) show one person; that person is the subject. Keep them instantly recognisable through what identifies them: face shape, eyes, eyebrows, nose, mouth and smile, hairstyle and hair colour, skin tone, glasses, beard or freckles, and their outfit with its colours and patterns. Their real body proportions do NOT carry over: head size, body, arms and legs follow the style below exactly, even where that is far from the photo. Do NOT copy the face from the photo: re-create the whole head, face, hair and neck from scratch in exactly the same style, material, level of detail, lighting and colour grading as the rest of the character, as if one artist made the entire character at once. The head sits naturally on the body at the character's proportions, with no photographic skin texture, no photo lighting, no visible seam, edge or colour shift at the neck, and no head that looks pasted on. If several photos are provided they all show the same subject from different angles or moments: merge them into ONE consistent figure and pick the clearest features. The photos may be cropped or show only the face or upper body: reconstruct the COMPLETE body anyway, from the top of the head to the feet or paws, including legs, feet, shoes, tail and all limbs, complete and believable at the proportions of the requested style, matching everything visible. Never crop, cut off or hide any part of the body, never show a bust or half figure, and do not leave anything outside the frame. Style: a premium designer vinyl collectible toy. Proportions: the head is about one third of the total height, slightly larger hands and feet, a compact body. Everything is simplified into soft, rounded, smooth shapes with no fine texture: hair as a few big sculpted chunks, clothing as clean simple shapes with only the main folds. Face: simple glossy painted eyes with a small white highlight, a small simple nose, a clean friendly smile. Flat, clean, slightly saturated colours taken from the outfit, smooth semi-gloss vinyl finish. Base: a plain round glossy white base. Pose: cheerful standing pose, hands on hips. Transform the subject into a single collectible figurine: the entire figure visible from head to toe with generous empty margin above the head and below the base, standing upright on a small, plain, round, flat base, perfectly centered, facing the camera straight-on in a frontal full-body view, with balanced margins on all sides and nothing cropped. Chunky, sturdy proportions with no thin fragile parts. Even, soft studio lighting with no harsh shadows. Isolate cleanly on a flat, seamless, pure white (#FFFFFF) studio background: no other objects, people, props, text, logos or shadows on the background, and clean accurate edges without halos. A sharp, clean product render of the physical figure. Most important: the whole figure, head and face included, is unmistakably in the stylized look described above, as one consistent physical object, with no photographic part.
"""

3. Build the 3D model
Call generate_prism_3_1 with the task_id of the image from step 2 in images, texture_model "3.5", texture_quality "fast", geometry_quality "standard", pbr false.

4. Export
Export it for printing: call convert_model on the 3D model with output_format "3mf" (or "stl" if I ask for it), plus max_size and unit "cm" if I gave you a print size.

After each generation call, keep polling get_generation with its task_id until it completes, then show me the result with show_generation before you start the next step.