listings_fetch_attributes_by_asins
Bulk fetch + flatten Amazon catalog attributes for N ASINs in ONE call.
readAmazon Seller (SP-API)sp.catalogItems_2022-04-01
Amazon SP-API · getCatalogItem · GET /catalog/2022-04-01/items/{asin} Amazon SP-API · getDefinitionsProductType · GET /definitions/2020-09-01/productTypes/{productType} For each item, fetches the catalog item, infers productType (or uses the explicit hint), looks up the schema, and flattens the whole JSON into cells keyed by Amazon's own flat-file keys — the operator can paste them straight into a Publish Queue row. To edit a listing you own, fetch by SKU (listings_fetch_attributes_by_skus) — an ASIN returns Amazon's merged catalog values, which can include other contributors' data; a full UPDATE would submit them as yours.
Call it
In an AI client (MCP)
{
"name": "listings_fetch_attributes_by_asins",
"arguments": {
"store": "myStore-US",
"items": []
}
}From code (Tool API)
curl -X POST https://api.sellersheetai.com/api/v1/tools/listings_fetch_attributes_by_asins \
-H "Authorization: Bearer $SELLERSHEET_API_KEY" \
-H "Content-Type: application/json" \
-d '{"store":"myStore-US","items":[]}'Early access — the Tool API is open on request, contact us.
Arguments
| Name | Type | Description |
|---|---|---|
storerequired | string | the store ref `<store>-<CC>` (name-country, e.g. 'myStore-US') — one of get_user_context's store_refs, verbatim. |
itemsrequired | object[] | List of {asin: str, productType?: str}. productType is auto-inferred from the catalog if omitted. |
locale | string | schema locale for the labels; defaults to the marketplace's. Allowed: DEFAULT | ar | ar_AE | de | de_DE | en | en_AE | en_AU | en_CA | en_GB | en_IN | en_SG | en_US | es | es_ES | es_MX | es_US | fr | fr_CA | fr_FR | it | it_IT | ja | ja_JP | nl | nl_NL | pl | pl_PL | pt | pt_BR | pt_PT | sv | sv_SE | tr | tr_TR | zh | zh_CN | zh_TW.DEFAULTarar_AEdede_DEenen_AEen_AUen_CAen_GBen_INen_SGen_USeses_ESes_MXes_USfrfr_CAfr_FRitit_ITjaja_JPnlnl_NLplpl_PLptpt_BRpt_PTsvsv_SEtrtr_TRzhzh_CNzh_TW |
language_tag | string | which language's entries to keep; defaults to the marketplace's primary listing locale. |
Result
Every tool answers {notification, data, human_action}; data carries:
data.rows— [{asin, productType, attributes: {key: cell},data.rows.names— {key: display_name},data.rows.data.rows.names.dropped_entries— {attr: sentence}} | {asin, error}] `attributes` keys are self-describing (Amazon's flat-file grammar); `names` gives a NEW key's display name (for a column the target sheet doesn't have yet); `dropped_entries` names, per attribute, any entry skipped for a foreign marketplace/language, in plain English. A row carries `note` when its product-type schema could not be served and the cells were typed from the JSON alone.
Params (Amazon schema; getCatalogItem)
asin*: string
marketplaceIds*: [string]
includedData: attributes | classifications | dimensions | identifiers | images | productTypes | relationships | salesRanks | summaries | vendorDetails (list)
locale: stringParams (Amazon schema; getDefinitionsProductType)
productType*: string
sellerId: string
marketplaceIds*: [string]
productTypeVersion: string
requirements: LISTING | LISTING_PRODUCT_ONLY | LISTING_OFFER_ONLY
requirementsEnforced: ENFORCED | NOT_ENFORCED
locale: DEFAULT | ar | ar_AE | de | … | zh_TW (38 values)
parentageLevel: NONE | CHILD | PARENTWorkflow
- 1. Read column matching `externally_assigned_product_identifier` (or `merchant_suggested_asin`) for the selected rows. 2. Call this tool ONCE with all ASINs. 3. Never rename or move a template column (the template is the source of truth). A bare key, `#1.value`, line 1 of `#*.value` and a literal marketplace/language bracket equal to row 3 are the same column — the template's spelling wins. 4. Any other key that is not yet a column — a leaf the template's productType schema doesn't have (the row's productType differs from the template's) — is APPENDED AT THE RIGHT END, named from `names`, so the seller sees the whole listing. Do not write a special row-4 note for it: its place says it was appended. Only a REPEAT of a cluster the template already has (entry index >1 whose `#1` twin is on the sheet) is inserted beside it. EVERY column is sent on submit — the sheet is trusted; whoever owns the sheet removes a column Amazon no longer takes after reading the result. 5. Write each row's cells back under their row-5 keys (attribute columns start at F; E is attributes_json) and set column D (productType) from the response.