Cross-Platform Product Price Comparison with One AI Agent
Let one AI agent compare the same product across 1688, Taobao, JD, Temu, Amazon and Shopee in a single instruction: it pulls price, sales and MOQ together, aligns the fields, flags the lowest price and best value, and outputs a cross-platform comparison table. No per-API integration.

Tool names in this guide are for reference only and may have changed or been renamed. Please use the latest tools in the tool marketplace when implementing.
Key takeaways
- One natural-language instruction lets a single agent search the same item across platforms and build one comparison table
- The agent aligns title, price, sales and MOQ, then flags the lowest price and best value
- Orchestration you can't get by wiring each platform's API yourself — one key handles all of it
Tools used
Why cross-platform price comparison is so hard
To know where a product is actually cheapest, in theory you just look up its price on 1688, Taobao, JD, Temu, Amazon and Shopee. In practice that's six tabs to juggle, six search boxes to fill, six currencies and units to reconcile, then everything copied into a spreadsheet by hand — is that a unit price or a wholesale price? Is "sales" monthly or lifetime? Are the minimum order quantities even in the same ballpark? Aligning those fields is more work than the lookups themselves, and one category eats half a day.
If you want software to do it, the traditional path means integrating each platform's API, maintaining separate sessions, and fighting anti-scraping — a bar too high to clear for a one-off comparison.
AgentEarth takes a different approach: each platform's search capability is packaged as a ready-made tool. You describe the goal in one natural-language instruction, and a single agent picks the tools, searches the same item on each platform, and aligns the fields for you. One agent, one instruction, many platforms — that's orchestration you can't get by wiring up each API on your own.
The workflow in five steps
- Give the target — hand the agent a product name or model.
- Search the same item across platforms — the agent queries each marketplace.
- Align the fields — put title, price, sales and MOQ into one table with one consistent definition.
- Flag the highlights — mark the lowest price and the best value.
- Output the table — produce one cross-platform comparison table.
How to do each step
Step 1: Name the target product in one line
Just tell the agent, e.g. use ae to compare the price and sales of a given wireless-earbuds model across the major marketplaces. The agent breaks that into "which platforms, what to search, which fields to align" — no API knowledge required.
Step 2: One agent searches each platform for the same item
This is the key orchestration step — the same agent calls one search tool per platform:
- Wholesale/sourcing:
ae_ali1688_product_search(1688). - Taobao/Tmall listings via an item-detail tool,
ae_tmapi_taobao_item_detail(Taobao/Tmall). - Domestic ecommerce:
ae_jd_search_products_v1(JD). - Cross-border value platform:
ae_temu_product_search_from_name(Temu). - Major overseas marketplace:
ae_amazon_product_search(Amazon). - Southeast Asia:
ae_shopee_product_search(Shopee).
You give one instruction; the agent fans these tools out across platforms for you.
Step 3: Align the fields
Each platform reports fields differently, so have the agent normalize them: collapse titles to the same item, put price into one consistent column scheme (unit vs. wholesale split out), and give sales and MOQ their own columns. Field alignment is the real difficulty in cross-platform comparison — let the agent handle it cleanly in one pass.
Steps 4 and 5: Flag highlights, output the table
On the aligned data, have the agent flag the lowest price and the best-value pick (weighing price, sales and MOQ together), then compile one comparison table:
| Platform | Backing tool | Price basis | Sales | Minimum order |
|---|---|---|---|---|
| 1688 | ae_ali1688_product_search |
Wholesale unit | High | Higher |
| Taobao/Tmall | ae_tmapi_taobao_item_detail |
Retail unit | Mid-high | 1 unit |
| JD | ae_jd_search_products_v1 |
Retail unit | Mid | 1 unit |
| Temu | ae_temu_product_search_from_name |
Retail unit | Mid | 1 unit |
| Amazon | ae_amazon_product_search |
Retail unit | Mid | 1 unit |
| Shopee | ae_shopee_product_search |
Retail unit | Mid | 1 unit |
(The bases shown are illustrative; actual numbers come from what the agent returns.)
How to get started
- Sign up for AgentEarth — new users get 1,000 free credits.
- In agents like WorkBuddy or Trae, search for "AgentEarth" and connect in-app; or paste a one-line MCP config into Claude Code / Cursor / VSCode.
- Give instructions in natural language, e.g. use ae to compare the same item's price and sales across marketplaces.
FAQ
Which platforms are supported? Currently 1688, Taobao/Tmall, JD, Temu, Amazon and Shopee, each backed by its own search tool (such as ae_ali1688_product_search, ae_jd_search_products_v1, ae_amazon_product_search). One agent can call several in a single run.
How are the fields aligned? Price basis, sales definitions and MOQ differ per platform, so the agent normalizes them into one table: price split into unit/wholesale columns, sales and MOQ in their own columns, and titles collapsed to the same item — so you never compare a wholesale price against a retail one.
Can it run a scheduled price-drop monitor? Yes. Hand the same comparison instruction to the agent on a schedule, and it periodically pulls each platform's price for the same item, compares against the previous run, and alerts you on a drop or stockout.
