---
name: ghost-mannequin-shot
description: Turn a hanger, flat-lay or mannequin photo of a garment into a clean invisible-mannequin product shot on white, symmetrical and filled out, with the inner neck label visible. The garment's colour, fabric and details stay the same.
metadata:
  type: image
  version: 1.0.0
  category: Fashion
  tagline: Use this Skill to turn a garment photo into a clean ghost mannequin shot for your shop.
  made-for: Fashion brands, E-commerce sellers, Marketplace sellers, Photographers
  generates-with: Still image, 4:5 or 1:1
  requires: Node.js 18 or newer, only to run scripts/build-input.mjs
  outside-costs: None in the Skill. The image model you send the request to may charge for each image.
  updated: 2026-10-16
  title: Ghost mannequin shot
  availability: live
  summary: Turn a garment photo into a clean invisible-mannequin shop image on white, with the garment kept exactly as it is.
  command: /ghost-mannequin-shot my olive bomber jacket on a hanger
  output: PNG · 4:5 or 1:1 on white
  required-input: Garment photo (hanger, flat-lay or mannequin)
  creator: meleon
---

# Ghost mannequin shot

Use this skill when someone has a quick photo of a garment on a hanger, on a bed or on a mannequin and needs the clean, hollow-form image shops expect: the garment standing on its own, as if worn by an invisible body.

## What you need

- One garment photo, as a public URL. The whole garment should be in frame.
- The garment type (jacket, shirt, dress, hoodie, trousers).
- The background: white (default) or light grey. Optional.
- Any detail that must be kept, for example "silver zip, sleeve pocket, ribbed cuffs". Optional.

If the photo is missing, ask for it. Do not invent the garment.

## What you make

One still image of the same garment in the invisible-mannequin style:

- Portrait 4:5 by default.
- Square 1:1 for marketplaces that require it, when asked.

## How to run it

1. Read `references/ghost-mannequin-guide.md`. Note which parts of the garment are visible in the photo and which must be rebuilt, such as the inside back neck.
2. Fill in `assets/prompt-template.md`, or run `scripts/build-input.mjs` to print the full request.
3. Send the request like this: POST the printed JSON `input` to fal model `openai/gpt-image-2/edit` with the user's own FAL_KEY (the Skill never stores keys), or paste the prompt and images into any GPT-Image-class editor. Use the garment photo as the reference image. The defaults (1600 × 2000, quality medium, about $0.10 per image) are the settings that made `assets/preview.webp`.
4. Check the result against `references/output-spec.md`: symmetry, hollow neck and sleeves, fabric and hardware unchanged, no hanger or mannequin left.
5. If a check fails, try again, at most 3 attempts in total. Change one thing per attempt, using the fix table in the guide, and name what you changed. View each output beside the input photo. Then return the best one and say what still differs from the garment.
6. Return the image and the prompt you used.

## Privacy

A photo uploaded to a public URL can be fetched by anyone who has the link. Delete the uploaded photos once the image is made.

## Files

- `references/ghost-mannequin-guide.md`: how the hollow look works, lighting and fixes.
- `references/output-spec.md`: sizes, format and the pass/fail checks.
- `assets/prompt-template.md`: the prompt to fill in.
- `assets/generation-input.json`: the model and default image settings.
- `assets/example-input.webp`: a sample hanger photo.
- `assets/preview.webp`: the result.
- `scripts/build-input.mjs`: prints the full model request. It does not call any API.