E-commerce

JD.com Product Search Data for AI Agents

Let one AI agent search JD.com products by keyword and pull back price, shop, and sales fields in bulk, then sort and organize them into an analyzable structured table. No login, no per-API integration, and candidates hand off to cross-platform price comparison.

JD.com Product Search Data for AI Agents

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

  • Search JD.com by keyword from one natural-language instruction and pull back price, shop, and sales in bulk
  • Results come back as structured fields you can sort, filter, and tabulate — not copy-pasted web pages
  • Candidates hand off directly to cross-platform comparison on other marketplaces, no new tools

Tools used

Why pulling JD.com data by hand is a grind

Getting a batch of JD.com product data for price comparison or product research is painful by hand: type a keyword, scroll several pages, open each listing to read price and shop, copy the promising ones into a spreadsheet, and keep straight the same-name-different-item cases and the gap between promo price and final price. One category eats most of a day — and comparing by eye, you routinely miss a cheaper or higher-selling listing.

If you want a program to do it for you, the traditional path means wiring up JD's data endpoints yourself, maintaining a session, and handling pagination and anti-scraping. That's a high bar and a heavy thing to maintain.

AgentEarth takes a different approach: these capabilities are packaged as ready-made tools. You describe the goal in one natural-language instruction, and the agent picks the tools, runs them, and organizes the results as structured data.

The workflow in four steps

  1. Give keywords and filters — tell the agent what to search, the price range, and how to sort.
  2. Search products by keyword — pull back a batch of candidate listings.
  3. Extract price, shop, and sales fields — pull the structured fields out.
  4. Sort and build a comparison table — have the agent rank by price or sales and output an analyzable table.

How to do each step

Step 1: State your data goal in plain language

Just tell the agent, e.g. use ae to search JD.com for a category, within a given price range, sorted by sales high to low. The agent breaks that into "which keyword to search, how to filter, how to rank" — no API knowledge required.

Step 2: Search JD.com products by keyword

Use a JD product-search tool (such as ae_jd_search_products_v1) to pull back a batch of candidate listings by keyword. This step returns a structured list rather than a pile of web pages, which makes everything downstream easy to process.

Step 3: Pull back price, shop, and sales fields

The same tool's response (ae_jd_search_products_v1) carries fields like price, shop, and sales. Have the agent extract and align these fields so you never click into each listing to verify them one by one.

Step 4: Sort and compile a comparison table

Have the agent sort by price or sales, drop the obvious misses, and compile a comparison table. Once you have candidates, the same agent can continue to other marketplaces to search the same item for a cross-platform comparison.

Approach Time Output form Analyzable right away?
Browse JD by hand Most of a day per category Scattered across pages / manual spreadsheet Needs manual cleanup before you can compare
Agent structured fetch One instruction, minutes A batch of structured price/shop/sales fields Sort, filter, and tabulate directly

How to get started

  1. Sign up for AgentEarth — new users get 1,000 free credits.
  2. 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.
  3. Give instructions in natural language, e.g. use ae to search JD.com for a category and build a comparison table.

FAQ

What fields come back? A JD product-search tool (such as ae_jd_search_products_v1) returns structured fields like product title, price, shop, and sales, which the agent can extract and align — no page-by-page copying.

Can it sort by price? Yes. Once you have the structured fields, have the agent sort and filter by price, sales, or any other field and output a ranked comparison table directly.

Can it compare prices across platforms? Yes. With JD candidates in hand, the same agent can continue to Taobao, 1688, Amazon and others to search the same item for a cross-platform comparison — no new tools and no re-integration.


Browse tools · Integration guide