Ecommerce Size Images

When to Use

Use this skill when the user wants marketplace-ready ecommerce size, dimension, or specification graphics from product photos plus confirmed product measurements. Typical requests include:

Create Taobao, Tmall, 1688, Amazon, Shopee, Temu, JD, Pinduoduo, or Xiaohongshu product size images.
Turn a product photo into a dimension diagram with width, height, depth, strap length, capacity, weight, package size, or SKU variant measurements.
Create a parameter/specification image, size chart, capacity diagram, comparison scale image, or annotated dimensions image.
Use reference images for layout, measurement-line style, table style, callout style, or platform visual tone.

Use this skill for factual measurement graphics. Use ecommerce-main-images for cover images and ecommerce-selling-point-images for benefit callout images. If no measurements or product parameters are supplied, do not fabricate them; ask for the minimum needed facts or produce a layout draft with clearly marked placeholders only if the user explicitly wants a draft.

Inputs

Required: at least one original product image. Preserve product identity, shape, proportions, color, material appearance, logo placement, packaging, SKU details, and distinctive marks.
Required for final deliverables: confirmed measurements or parameters. Accept any structured form the user provides, such as 宽 32cm，高 28cm，厚 12cm, 肩带长 58cm, 重量 0.6kg, 容量 12L, S/M/L size table, package dimensions, or variant dimensions.
Optional: reference images. Use them only for layout, callout style, measurement-line style, table styling, background, or typography. Do not copy unrelated products, competitor branding, watermarks, or exact ad layouts.
Optional: target platform. Default to a clean domestic ecommerce spec style when unspecified. Amazon-style graphics should be plain, factual, and restrained; Taobao/Tmall/1688 can use richer layout; Xiaohongshu can be more editorial but still factual.
Optional: target language. Default to the user’s conversation language.
Optional: unit system. Preserve user-supplied units. Do not convert units unless requested; if converting, keep both original and converted values when useful.
Optional: image type. Default to dimension diagram. Other modes: spec table, size chart, capacity diagram, package size, variant comparison, or detail + dimension hybrid.
Optional: aspect ratio or output size. Default to square 1024x1024; use 1024x1536 for vertical detail/spec graphics and 1536x1024 for wide comparison tables.
Optional: output count. Default to 1. Pass n only when the user asks for multiple variations.
Optional: output format and quality. Default to PNG and quality: "high" for final deliverables.

Treat measurements as a source-of-truth table before prompting. Separate:

confirmed: supplied by the user or visible as printed labels/packaging.
visible but unmeasured: facts visible in the image, such as shape, handle count, compartments, pattern, or included accessories.
missing: values needed for the requested graphic but not supplied.
do not claim: material, capacity, weight, durability, waterproofing, certifications, size fit, or compatibility unless supplied.

Execution

Use $gpt-image-2 for the actual image operation. This skill provides ecommerce size-image strategy, fact handling, prompt structure, product preservation rules, and platform constraints; $gpt-image-2 provides the execution path.

Build the fact sheet before generating.

Do not infer exact measurements from the photo. If the user asks for a final size image and required values are missing, ask one concise follow-up for the missing measurements. If the user wants a draft without confirmed values, use explicit placeholders such as 宽度：待确认 and state that it is a draft.

Follow $gpt-image-2 image editing mode.

Because this workflow starts from real product imagery, use GPT Image Editing through fusion-api.openai_image_edit_async_submit, as documented by $gpt-image-2. Do not run new oo capability discovery during normal use.

Upload local images that must be passed to $gpt-image-2:

oo file upload "<product-image-path>" --json
oo file upload "<reference-image-path>" --json

Pass each returned downloadUrl as an image reference in images, using the $gpt-image-2 shape:

{"image_url":"<downloadUrl>"}

Put primary product images first, then optional layout/style references.

Choose the size-image mode.

dimension diagram: product hero with measurement arrows/lines for width, height, depth, strap length, or other confirmed dimensions.
spec table: clean table of confirmed product parameters such as material, color, size, weight, package, or model only when supplied.
size chart: apparel or variant table, useful for S/M/L, waist, bust, length, shoulder, foot length, or ring sizes.
capacity diagram: show what fits or capacity only when supplied or when contents are explicitly provided by the user.
package size: packaging dimensions, carton size, gross weight, or shipping info only when supplied.
variant comparison: compare colors, sizes, or SKUs only when supplied or visible.

Build a concise prompt with this structure:

Create a marketplace-ready ecommerce size/specification image.
Primary product: preserve the product from image 1 accurately, including shape, proportions, color, material appearance, packaging, labels, logo placement, SKU details, and distinctive physical features.
Reference usage: use images 2+ only for measurement-line style, table layout, typography, background, or composition inspiration; do not copy unrelated products, competitor brands, watermarks, platform badges, or text.
Platform: <platform or general ecommerce>.
Target language: <language>.
Image type: <dimension diagram | spec table | size chart | capacity diagram | package size | variant comparison>.
Confirmed data to display exactly: <structured measurements and parameters>.
Missing data: <omit from final image, or show as 待确认 only if user requested a draft>.
Composition: product-first, clear measurement arrows and labels, mobile-readable text, tidy grid/table when needed, factual layout, enough whitespace, professional ecommerce visual hierarchy.
Avoid: invented numbers, invented material, invented capacity, fake certifications, fake weights, fake sizes, QR codes, phone numbers, social handles, external URLs, watermarks, platform badges, unsupported claims, distorted product details, clutter, unreadable text.

Execute with $gpt-image-2.

The underlying command shape is:

oo connector run "fusion-api" \
  --action "openai_image_edit_async_submit" \
  --data @payload.json \
  --json

Example payload:

{
  "model": "gpt-image-2",
  "prompt": "Create a marketplace-ready ecommerce size/specification image...",
  "images": [
    {"image_url": "<product-downloadUrl>"},
    {"image_url": "<optional-reference-downloadUrl>"}
  ],
  "output_format": "png",
  "quality": "high",
  "size": "1024x1024"
}

Use exact $gpt-image-2 field names: output_format, quality, size, and optional n. oo connector run --json usually waits internally and returns completed image results directly; if it returns a handle, follow $gpt-image-2 polling guidance.

Result Handling

Read GPT Image 2 results according to $gpt-image-2:

Image URLs are usually in .data.data[].url.
Revised prompts may be in .data.data[].revised_prompt.
Returned metadata may include .data.size, .data.quality, .data.output_format, .data.model, and .data.usage.

Download each returned HTTP image URL with oo file download "<url>" "<output-dir>" --name "<fileNameWithoutExtension>" --ext "<png|jpeg|webp|jpg>". oo file download prints Saved to: <path> and does not support --json.

Use a clear local output directory such as /Users/yunshi/Downloads/ecommerce-size-images/<short-product-name>-<timestamp>/.
Name files predictably, for example size-image-01.png, spec-image-01.png, or size-chart-01.png.
Preview the generated image when practical, or return the local file path plus the confirmed data used.
In the response, mention if any requested measurement was omitted because it was missing or unconfirmed.
If the task returns structured result data instead of direct URLs, report the actual returned fields and do not invent file URLs.

Failure Handling

Missing original product image: stop and ask for one product image.
Missing measurements for a final size image: ask for the minimum required values instead of generating fake numbers.
User asks to estimate dimensions from the photo: refuse exact estimation; offer a layout draft with 待确认 placeholders or ask for confirmed measurements.
Conflicting measurements: ask which source is authoritative before generating.
Too many reference images: select the most relevant references for measurement layout/style and summarize the rest in the prompt.
Unsupported size, output format, quality, or count: use the nearest $gpt-image-2 supported value and mention the adjustment.
User asks for proof-dependent claims: omit material, capacity, waterproofing, load-bearing, certifications, safety, compatibility, or regulatory claims unless supplied.
Upload failure: report the exact file that failed and retry only after the path or network issue is resolved.
Connector handle or timeout: follow $gpt-image-2 result/polling guidance before rerunning. Do not start a duplicate task only because a wait window ended.
GPT Image 2 connector failure or billing/auth blocker: report the exact blocker from oo output and the next useful action.

Ecommerce Size Images

SKILL.md

Ecommerce Size Images

When to Use

Inputs

Execution

Result Handling

Failure Handling

Выберите настройки cookie