alibaba1688_list_offer_variants

Read every variant of one 1688 product and store them for the account, so each can be paired; then call alibaba1688_pair_sku with a variant's item_id.

writealibaba1688

Call it

In an AI client (MCP)

tools/call
{
  "name": "alibaba1688_list_offer_variants",
  "arguments": {
    "account": "<account>",
    "offer": "<offer>"
  }
}

From code (Tool API)

curl -X POST https://api.sellersheetai.com/api/v1/tools/alibaba1688_list_offer_variants \
  -H "Authorization: Bearer $SELLERSHEET_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"account":"<account>","offer":"<offer>"}'
Early access — the Tool API is open on request, contact us.

Arguments

NameTypeDescription
accountrequiredstringthe 1688 account ref (its 1688 login name) — one of get_user_context's alibaba1688 account_refs, verbatim.
offerrequiredstringthe 1688 product link `https://detail.1688.com/offer/<id>.html`, or the offer id alone (digits). A short link is refused.

Result

Every tool answers {notification, data, human_action}; data carries:

  • data.offer_id / data.subject — the product read.
  • data.items[] — one per variant, and one more per further local SKU it is paired with — item_id, spec_id, attributes [{name, value}], price_yuan, unit, min_order_qty, supplier_login_id, state, sku (the local SKU of that row, or null).

Gotchas

  • 1688 serves a product's detail only for a supplier the account has bought from, and meters it on a paid quota: a refusal carries 1688's own sentence; relay it.
  • A variant stored before that 1688 no longer lists becomes state gone; a paired one keeps its pairing and cannot be ordered until paired again.

On this page

Actions

View as Markdown

Ask an assistant

Ask ChatGPTAsk Claude