For AI

Answers

How do I get Xiaohongshu (RedNote) trending topics and notes by category or industry from outside China?

Short answer

Through okdata, from any country and with no Xiaohongshu account: the hot topics list costs $0.02 a call for about 20 topics, and PGY's trending notes $0.04 a call for about 34 notes with reads, likes and saves. No endpoint takes a category or industry parameter, so you narrow by words: PGY trending notes take a Chinese keyword such as 护肤 (skincare), and topic notes return the hottest notes under one hashtag page. A daily snapshot of hot topics plus the top notes under 5 hashtags costs $3.60 a month.

An agency in London sends a skincare client a note every Monday on what Chinese consumers are excited about. Xiaohongshu (RedNote) is where those trends show up first, but the app has no export and its hot topics list mixes every subject. The list for Nov 16 opened with an Egyptian-pharaoh ice cream, a fictional character's backstory and a TV drama plot; a Labrador knocking over a guzheng was sixth. None of its 20 entries was about beauty.

So "by category" is something you build. None of the endpoints below has a category or industry parameter. What works is a short list of Chinese words for your industry, the few hashtag pages your category gathers around, and an LLM to sort the general list.

The easy route: ask an agent

Connect once: add https://api.okdata.online/mcp as an MCP server in Claude, Cursor or another client (steps on the MCP page), or give your agent the one-sentence setup in agents.md. Then ask in plain words:

  • "Show Xiaohongshu's trending notes about 护肤 (skincare) from the last 7 days. Translate the titles and list reads, likes and saves."
  • "Get today's Xiaohongshu hot topics, translate them, and keep only the ones about beauty, food or travel."
  • "For the hashtag page in this link, get its views and its 10 hottest notes, and summarise them in English."

The agent picks the endpoint, writes the Chinese query and translates the answer. For an agent that runs on a server abroad, see Xiaohongshu data for AI agents outside China.

What each endpoint lets you narrow

Endpoint How to narrow it What comes back Price
Hot topics It can't be narrowed; tag the topics yourself About 20 topics with rank, heat and a search link $0.02
PGY trending notes query words, board, window (3d, 7d, 14d or 30d), type (image or video) About 34 notes a page with reads, likes, saves, comments and the creator's followers $0.04
Topic notes One hashtag page ID The 10 or so hottest notes under it, with full text $0.02
Topic detail One hashtag page ID The page's views and how many people it says are discussing it $0.02
Note search Words, sort=popular About 20 notes a page $0.02

PGY (蒲公英) is Xiaohongshu's marketplace where brands book creators, and its content square ranks trending notes. Its eight board values are commercial purposes, not industries: hot (the default), seeding for product seeding, shop_hot and shop_promo for e-commerce, pgy for sponsored notes, leads, direct_seeding and app_promo. For a consumer brand, query does the category work: 护肤 for skincare, 彩妆 for makeup, 咖啡 for coffee, 露营 for camping, 宠物 for pets. The words are matched against the whole note.

A month of one industry, priced

The job: a snapshot of the hot topics every day, the hottest notes under 5 hashtag pages every day, the views of those pages once a week, and PGY trending notes for 3 industry words once a week.

Task Calls Cost
Hot topics, once a day for 30 days 30 $0.60
Hottest notes under 5 hashtag pages, daily 150 $3.00
Views of the 5 hashtag pages, weekly 20 $0.40
PGY trending notes for 3 words, 7-day window, weekly 12 $0.48
The month 212 $4.48

You pay per call, with no subscription. Every endpoint has its own price, shown on its page, and a new key comes with $0.10 of credit.

Code

import os, requests

API = "https://api.okdata.online"
H = {"Authorization": f"Bearer {os.environ['OKDATA_API_KEY']}"}

def get(path, **params):
    body = requests.get(API + path, headers=H, params=params, timeout=60).json()
    if body["code"] != "ok":
        raise RuntimeError(f"{body['code']}: {body['message']}")
    return body["data"]["items"]

# 1. Trending notes for one industry word over the last 7 days
for n in get("/v1/xiaohongshu/pgy/notes/trending", query="护肤", window="7d",
             fields="title,url,metrics.views,metrics.likes,metrics.favorites,author.followers")[:10]:
    print(n["metrics"]["views"], n["metrics"]["favorites"], n["title"])

# 2. The hottest notes under the hashtag pages you follow (24-character IDs from hashtag links)
for topic_id in ["5ee1c0a110694c00010b4ae9"]:
    notes = get("/v1/xiaohongshu/topics/notes/list", id=topic_id, fields="title,text,url,metrics.likes")
    print(topic_id, len(notes), "notes")

# 3. Today's hot topics, for an LLM to tag by industry
topics = get("/v1/xiaohongshu/trends/list", fields="rank,name,heat")
print([t["name"] for t in topics])

Free, in English, without a key

  • The Xiaohongshu hot topics board shows today's list with each entry in English and the Chinese beside it.
  • The weekly digest picks the stories that stayed longest on China's hot lists, translated, with where they ranked.
  • The china-trends repository keeps a CSV of the Xiaohongshu list for every day since Nov 12, 2026, with a machine-translated name_en column; check the Chinese before quoting it. Give a month of it to an LLM with your category's definition and you have a category history without running a scraper.

Things that trip people up

  • Hot topics come back as names with a search link, not hashtag page IDs. To see the notes behind a hot topic, run note search on its name with sort=popular. Topic notes and topic detail need the 24-character ID from a hashtag's link, which you copy once for each hashtag you follow.
  • A PGY board never goes past its top 100 notes, with or without words, so a narrow word over 3 days may return only a few.
  • PGY trending notes carry reads, which public Xiaohongshu pages don't show, but only titles, no note text. Topic notes carry the full text and no reads.
  • Search in Chinese. 护肤 finds far more than "skincare".
  • Neither list keeps history. Store each day's answer with its meta.fetched_at time, or start from the CSVs.

A Chinese-speaking colleague with the app can scroll the same lists for free, which is enough for a one-off look. The data route pays off when you need the same view every week, in English, in a spreadsheet or a report.

Sources

Where people asked this (read 2026-11-18)

Updated

Connect once, then just ask.

Every new key comes with $0.10 to try it. One sentence connects your agent.

Start free
Connect your AI

Paste this into your agent

Read https://okdata.online/agents.md and set up okdata for me; my key is in the OKDATA_API_KEY environment variable.

Claude Code, Cursor, Codex and other agents that can open a link set themselves up from it.

MCP serverhttps://api.okdata.online/mcp