generate_image
Generate images from a text prompt — listing photos, A+ banners, lifestyle and infographic shots, icons.
writeSellerSheet
Priced per delivered image (credits_quote gives the number; HTTP 402 with error_code insufficient_credits when the balance is short — top up at sellersheetai.com/billing); a failure is refunded.
Call it
In an AI client (MCP)
{
"name": "generate_image",
"arguments": {
"store": "myStore-US",
"prompt": "<prompt>"
}
}From code (Tool API)
curl -X POST https://api.sellersheetai.com/api/v1/tools/generate_image \
-H "Authorization: Bearer $SELLERSHEET_API_KEY" \
-H "Content-Type: application/json" \
-d '{"store":"myStore-US","prompt":"<prompt>"}'Early access — the Tool API is open on request, contact us.
Arguments
| Name | Type | Description |
|---|---|---|
promptrequired | string | the image description. |
provider | string | 'openai' (default; gpt-image-2 — precise, legible text) | 'nanobanana' (Gemini — fast, search-grounded; renders and is priced at the default canvas, so size, quality and background apply to openai only) | 'both'.Default: "openai" |
size | string | '1024x1024' (default, the same canvas Studio uses) | '2048x2048' (Amazon zoom headroom; costs more) | '1024x1536' | '1536x1024' | 'auto'.Default: "1024x1024" |
quality | string | 'low' | 'medium' | 'high' (default) | 'auto'.Default: "high" |
n | integer | variations, 1–4 (default 1).Default: 1 |
background | string | 'auto' (default) | 'opaque'. gpt-image-2 has no transparent background — 'transparent' is treated as 'auto' and the response carries background_note; generate opaque and remove the background downstream.Default: "auto" |
output_format | string | 'png' | 'jpeg' | 'webp'; omit and the server decides (currently jpeg). |
use_search_grounding | boolean | Google Search grounding (nanobanana); default false.Default: false |
synthetic_performer | boolean | Amazon's contains-synthetic-performer disclosure (XMP metadata in the upload); omit and the server infers it from the prompt; true or false overrides. set_synthetic_performer toggles it later without regenerating. |
store | string | the store ref `<store>-<CC>` (name-country, e.g. 'myStore-US') — one of get_user_context's store_refs, verbatim. Required. |
parent_sku | string | the family SKU; defaults to sku. |
sku | string | the child SKU. Required. |
slot | string | the Image Library slot: 's0'–'s8' (PDP) or 'a1'–'a5' / 'p1'–'p5' (A+). Required — a call missing store, sku or slot is refused (400) before billing. |
filename_prefix | string | upload filename prefix.Default: "generated" |
Result
Every tool answers {notification, data, human_action}; data carries:
data.images— one entry per image — cdn_url (saved at store/parent_sku/sku/slot), thumbnail_url, provider, model, revised_prompt (how the model rewrote the prompt — mention it), background_note (relay it when set), or error. An entry with r2_upload_error has no cdn_url; its image arrives as an inline image block.data.credits.charged / data.credits.balance— this call's cost and what is left.