Upfront: I did not build this. It is an open-source project on GitHub (superjack2050/1688-cli, MIT). It is listed here on star count, maintenance activity and how well it fits day-to-day storefront operations.
What it does
- Keyword sourcing —
1688 searchsorts by best-selling and filters by price range, so you screen suppliers in the terminal instead of paging through the site. - Batch research datasets —
1688 researchruns several keywords at once and emits JSONL you can feed straight into a script or a model. - Supplier evaluation — pulls credentials and transaction data for side-by-side comparison rather than one product page at a time.
- Agent-shaped output — structured results and clear command semantics, so an AI agent can orchestrate it inside an automated sourcing flow.
How to use it
Install globally from npm; the first run asks you to log in by scanning a QR code with the 1688 app:
npm i -g 1688-cli
1688 doctor # check the environment
1688 login # scan the QR once
# keyword search
1688 search "phone case" --sort best-selling --price-max 50
# multi-keyword research, JSONL out
1688 research "phone case" "usb cable" --max-per-query 50 --jsonl
The session is stored locally for later commands. The JSONL output is ready to hand to an agent for filtering and scoring.
When to use it, and when not to
This is the supply-side half. Pair it with the product-research skills in eCommerce Skills for demand-side validation and you have the full loop.


