Douyin Product-Library Selection with AI Agents (Chanmama)
Let one AI agent find what's trending in Douyin's product library: search hot items by category or keyword, open each product's details, and pull the videos currently selling it. One chain from winning product to selling strategy — no manual leaderboard digging.

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
- Find trending items in Douyin's product library, check details, and pull related selling videos from one natural-language instruction
- Go beyond the sales leaderboard — follow a winning product straight to who is selling it and how
- Selection results come with a built-in selling angle you can hand off to creator analysis or content planning
Tools used
Why selection on Douyin is so hard
Douyin commerce moves fast — today's winner can go cold in two weeks. For a selection operator, watching the sales leaderboard is nowhere near enough: it tells you what is selling, not why it sells or who made it take off. So you end up scrolling categories in the product library, opening each listing to read specs and trends, then jumping to creator pages to guess which video drove the spike. A full loop later, you still have scattered facts that are hard to turn into an executable selling plan.
If you want an AI agent to run this for you, the traditional path means wiring up Douyin commerce data sources yourself, maintaining a session, and handling pagination. That's a high bar.
AgentEarth packages these capabilities as ready-made tools. You describe the goal in one natural-language instruction, and the agent picks the tools, runs them in order, and assembles the "winning product plus selling chain."
The workflow in four steps
- Give a target category — tell the agent where you want to source.
- Find trending items — pull hot candidates from the product library by category or keyword.
- Check details — open each candidate's product details to confirm it's the right item.
- See how it's sold — pull the related selling videos to learn what content drives it.
How to do each step
Step 1: State your selection goal in plain language
Just tell the agent, e.g. use ae to find trending items in a category in Douyin's product library, list the candidates, and tell me how they're being sold. The agent breaks that into "which category to search, how to filter for winners, which videos to pull" — no API knowledge required.
Step 2: Find trending items in the product library
- With a specific category in mind, use a category-sourcing tool (such as
ae_chanmama_product_category_search) to pull the items trending in that category. - For keyword or cross-category discovery, use a product-library search tool (such as
ae_chanmama_product_library_search) to search winners by term.
Either way the agent gets a batch of candidates carrying sales and trend signals, instead of you scrolling pages.
Step 3: Check product details to confirm the fit
For the candidates, use a product-detail tool (such as ae_chanmama_product_detail) to read specs, price band, and trend one by one, ruling out items that chart well but don't actually fit you.
Step 4: Pull related selling videos to see how it sells
Once you've shortlisted a few candidates, use a related-videos tool (such as ae_chanmama_product_related_videos) to pull the videos currently selling the item, and let the agent summarize their content patterns (selling angles, scenes, creator type). This step turns "what you picked" straight into "how to sell it."
| Approach | What you see | Decision completeness | Produces a selling plan? |
|---|---|---|---|
| Sales leaderboard only | What is selling | Only the outcome | You still guess how to sell |
| Winner + details + selling videos | What sells, whether it fits, who sells it and how | Product to content, connected | Selection comes with a selling angle |
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 find trending items in a Douyin category and pull their selling videos.
FAQ
Can it find winners by category? Yes. With a specific category, use a category-sourcing tool (such as ae_chanmama_product_category_search) to pull what's trending there; for keyword discovery, use a product-library search tool (such as ae_chanmama_product_library_search).
Can it show who's selling an item? Yes. For a shortlisted product, use a related-videos tool (such as ae_chanmama_product_related_videos) and the agent will pull the videos selling it and summarize their content patterns.
How fresh is the data? This workflow fetches data through AgentEarth's tools in real time. You only need an AgentEarth key and natural-language instructions — no Douyin session to maintain and no API to integrate yourself.
