Shopee Market Research for AI Agents
Let one AI agent map a Shopee Southeast Asia market end to end: search live listings by name or keyword, read category trends and category data, round it out with keyword demand, and output product-selection recommendations. One chain, no per-country tool hunting.

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
- Map a Shopee market from one natural-language instruction: search listings, read category trends, add keyword demand, output recommendations
- Search by product name or by keyword so hits track what is actually selling
- Selection results hand off directly to 1688 sourcing — research and procurement use the same agent
Tools used
Why Shopee research feels so fragmented
Shopee spans several Southeast Asian storefronts, and each market has its own bestsellers, category structure, and keyword habits. To understand a single category, you usually open each country's site, scroll through live listings one by one, jot down price bands, then guess which keywords carry volume. The data is scattered across pages, and stitching a full picture together is all manual — it's easy to miss a category that's quietly taking off.
If you want an AI agent to do this for you, the traditional path means wiring up each storefront's data yourself, maintaining a session, and handling pagination. That's a high bar.
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 in order, and organizes the results.
The workflow in four steps
- Set the target — tell the agent the target market and category.
- Pull live listings — search current listings by product name or keyword.
- Read category trends — read category trends and category data to see which direction is gaining.
- Add keywords, output recommendations — round out the keyword view and have the agent compile a selection recommendation.
How to do each step
Step 1: State the target market and category
Just tell the agent, e.g. use ae to look at bestsellers and category trends for a category on Shopee Southeast Asia. The agent breaks that into "what to search, which dimensions to read, how to rank" — no API knowledge required.
Step 2: Pull live listings by name or keyword
- With a specific product name, use a name-search tool (such as
ae_shopee_product_search_from_name) to hit matching live listings directly. - With only a category keyword, use a product-search tool (such as
ae_shopee_product_search) to pull a batch of candidates, with price bands and sales visible at once.
Step 3: Read category trends and category data
Use a category-trend tool (such as ae_shopee_category_trend) to see whether the direction is rising or falling recently, then a category-data tool (such as ae_shopee_category_request) to pull the category structure and detail, so you can judge which sub-category to bet on.
Step 4: Add keywords and compile a recommendation
Use a keyword tool (such as ae_shopee_keyword_search) to round out the search-demand view and see what shoppers actually type, then have the agent fold listings, trends, and keywords into one selection recommendation.
| Approach | Coverage | Data consistency | Hand off to procurement? |
|---|---|---|---|
| Check Shopee country by country | Whatever storefront you reach; easy to miss | Each page on its own terms; stitched by hand | Re-search manually on other platforms |
| Run this agent workflow | Listings + trends + keywords in one pass | One consistent output in one go | Hands off directly to 1688 sourcing |
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 research a category on Shopee and give a selection recommendation.
FAQ
Which Southeast Asian storefronts are supported? This workflow fetches data through AgentEarth's Shopee tools. You state the target market in natural language, and the agent pulls live listings and category data for the storefront and category you name — no need to maintain a session per country yourself.
Can it read category trends? Yes. Use a category-trend tool (such as ae_shopee_category_trend) to see direction, then a category-data tool (such as ae_shopee_category_request) to pull structure and detail, and judge the sub-category together.
Can it connect to 1688 sourcing? Yes. Once Shopee research settles which products to pursue, the same agent can continue to 1688 to source suppliers, check specs, and build a candidate comparison table — research and procurement stay in one tool. See the 1688 product sourcing workflow guide.
