Research China consumer trends from Xiaohongshu notes
Search Xiaohongshu (RED) notes by keyword and get them as JSON, to see what Chinese shoppers recommend and ask about before you launch or enter the market.
Before you launch a product in China or write copy for Chinese shoppers, you want to know what they already recommend to each other, and on Xiaohongshu that happens in notes: lists, reviews and how-tos, in Chinese, ranked by an app you cannot search from a script. Lovely Router runs the search. Send a keyword, get one JSON object per item, pay per item from a prepaid balance. Call it from a script, a research notebook or an AI agent with one API key.
Hand it to your AI agent
Copy this prompt into Claude Code, Cursor, Codex or any coding agent. It tells the agent where the docs are and how to use your API key.
Research China consumer trends from Xiaohongshu notes. Use the Lovely Router data tools: each returns JSON and is billed per result.
- Docs (read first): https://lovelyrouter.com/docs/tools/find.md and https://lovelyrouter.com/docs/tools/run.md
- Plan first: POST https://lovelyrouter.com/v1/tools/suggest with `{"task": "<my task in plain words>"}` (free). It returns the tools to run in order, each with an example `input`, the `input_required` fields, an estimated cost and a ready-made `run` request. Replace the example values with my task's; if a field is unclear, read the tool with GET https://lovelyrouter.com/v1/tools/<id>. To browse instead, POST https://lovelyrouter.com/v1/tools/discover searches the catalog.
- Show me the plan and its estimated cost (`total_est_cost_usd`) before running anything large.
- Run: POST https://lovelyrouter.com/v1/tools/<id>/run with `{"input": {...}, "max_results": N}`. A 200 has the finished run with `items`; a 202 means it is still running, so poll the `poll` path until `status` is `succeeded`, `failed`, `timed_out` or `stopped`.
- Only delivered results are billed. Keep max_results small on the first run, check the output, then scale up.
- If you support MCP, connect https://lovelyrouter.com/mcp instead (Streamable HTTP, same key; https://lovelyrouter.com/docs/tools/mcp.md) and use suggest_tools, get_tool and run_tool.
- Auth: read my API key from the LOVELY_API_KEY environment variable and send it as `Authorization: Bearer $LOVELY_API_KEY`. Never print or commit the key. If it is not set, ask me to create one at https://lovelyrouter.com/app/keys.
- Billing is prepaid: every call is charged to my balance, so test with a small request before running anything large.
- Every docs page is plain Markdown; the full index is https://lovelyrouter.com/llms.txt.
My task: <the inputs (links, keywords or accounts) and how many results you need>How it works
Pick keywords in Chinese
Use the words shoppers use: a product type, an activity, a problem. Start broad, such as a hobby, then narrow.
Run the tool with a cap
Send the keyword to
POST /v1/tools/xiaohongshu-search/runwithmax_results. The run holds its maximum cost from your balance and returns the notes as JSON.Read for patterns
Sort by what is most liked, translate with a model, and look at repeated brands, price points and complaints.
Search notes with one request
Search Xiaohongshu notes takes an operation, here search_notes, plus a keyword and a sort_type: general, time_descending, like_count_descending, comment_count_descending or collect_count_descending. This tool has no cap field in input; the cap is max_results in the request:
curl https://lovelyrouter.com/v1/tools/xiaohongshu-search/run \
-H "Authorization: Bearer $LOVELY_API_KEY" \
-H "content-type: application/json" \
-d '{
"input": {
"operation": "search_notes",
"keyword": "露营",
"sort_type": "like_count_descending"
},
"max_results": 30
}'A run still going after 60 seconds returns 202; Run a tool shows how to poll.
What each item holds
| Field group | What you get |
|---|---|
| Notes | Xiaohongshu (RED) notes that match your keyword |
| Other operations | The same tool also takes search_products, get_note_detail, get_note_comments, list_user_notes, get_user_info, list_topic_notes, search_hot_list and others, each with its own fields |
The tool is described as notes for a keyword, so check the shape of a small run before you build a pipeline on specific fields. Text comes back in Chinese.
Recipes
Search the category keyword sorted by likes, translate the titles with a model, and list the brands and products that keep coming up.
Collect 100 notes on a problem, such as dry skin or small-space storage, and ask a model for the phrases shoppers repeat. Use them in listings and ads.
Search a seasonal keyword with time_descending each week and watch which sub-topics rise.
Take the note id of a high-performing note and run get_note_comments to read what people asked under it.
Use search_hot_list as a daily snapshot of what people are looking at.
Prefer plain words? Connect the MCP server and tell Claude Code or Cursor: "Search Xiaohongshu for 露营, sorted by likes, translate the titles to English and group the notes into five themes. Set max_results to 50."
Pair it with other Chinese sources
Notes show what people say. Marketplaces show what they buy:
Xiaohongshu (RED) notes for a keyword.
Douyin videos for keywords.
Weibo posts for a keyword.
Taobao and Tmall product pages by link: title, price and promotions, shop, brand, images, specs, sales and variants.
Second-hand listings from Xianyu (Goofish, China's largest resale market) by keyword: title, price and original price, condition, city, seller, photos, posted date.
douyin-searchcovers the short-video side of the same topic.weibo-searchshows public discussion on Weibo.taobao-productgives prices and sales for products the notes recommend.xianyu-listingsshows resale prices for the same goods.
For the wider setup, see Chinese social media data.
Notes and comments come from real users. If you store or contact them, you are responsible for using that data lawfully, including privacy and anti-spam rules.
What you pay
You pay per item: starting a run holds its maximum cost, the per-call fee plus the per-item fee times max_results, and the run settles to what it returned. The rest of the hold goes back to your balance. A run that fails or times out pays only for the items it delivered, and a run refused at the start costs nothing.
Pulling 50 items is billed as 50 items. There is no subscription and no monthly minimum: top up from $5, and one balance pays for tools and models. The card above shows today's price; pricing lists every tool.
Frequently asked questions
Do I need to search in Chinese?+
Chinese keywords match how users write. The example in the data uses 露营 (camping).
How many items can one run return?+
Up to 100, and 10 by default. Set max_results to the number you want.
Can I read comments or a creator's notes?+
Yes, by changing operation: for example get_note_comments, list_user_notes or get_user_info, with the matching note_id or user_id.
Which fields does a note contain?+
The tool describes the result as notes for a keyword and does not list fields. Run a small test to see them.
Do I need a Xiaohongshu account?+
No. You send a keyword and an API key.
Do I need a subscription?+
No. You add credit to a prepaid balance and each run draws it down. Failed runs pay only for the results they delivered, and nothing renews automatically.
Working across Chinese platforms? See Chinese social media data, influencer research and the rest of the tool catalog.