Xiaohongshu Scraper: RedNote (XHS) notes, profiles and trends

This Xiaohongshu scraper (also called RedNote, XHS or Little Red Book, 小红书) turns public notes into JSON, CSV or Excel, with no login or cookies. Search by keyword (Chinese terms such as 咖啡, 护肤, 露营 work best) or paste note links, and get title, text, hashtags, likes, saves, comments, shares, images, the video link and the author.

Add profile links for followers, notes posted and total likes, or read the home feed and topic channels. Used for China market research, trend tracking and creator discovery. Full notes from $6 per 1,000.

At a glance

• Full note (from a link, or a feed note opened): US$ 6.00 per 1,000
• Note card (feed or profile, not opened): US$ 3.50 per 1,000
• Profile: US$ 20.00 per 1,000
• Fee per run start: None
• Inputs: Keywords (best in Chinese), note links (with their xsec_token, or without it for notes indexed by search engines), profile links or ids, home feed and topic channels
• Error rows (expired link, missing token, profile not found): Free, with an errorCode

One real row

A note link from a cloud run of 2026-09-30:

{
  "type": "note",
  "id": "6455b3d7000000000800e27f",
  "url": "https://www.xiaohongshu.com/explore/6455b3d7000000000800e27f?xsec_token=ABj6wOII3wKf1doBy-UiXlhHC-HP-A9ammCgjhvhiT1L8%3D&xsec_source=pc_feed",
  "noteType": "normal",
  "title": "听我说！苏州平江路要早上7点就去，美的嘞",
  "desc": "Hi，大家好，我是逃喵喵🍃。\n\t\n我来过苏州啦！！好激动好激动呀，平江路被称为是“苏州古城的缩影”...",
  "tags": ["翻翻上周好天气", "阳光是最好的滤镜", "苏州", "苏州平江路", "平江路", "苏州旅游", "..."],
  "likedCount": 1619,
  "collectedCount": 1323,
  "commentCount": 151,
  "shareCount": 502,
  "publishedAt": "2023-05-06T01:56:39Z",
  "imageCount": 14,
  "images": [
    "https://sns-webpic-qc.xhscdn.com/202610010452/c455a6de62d55f1b94054fdb8a89f838/1000g0082etmn6f4h40005o2pg4ag9ap9ig46b28!nd_dft_wlteh_webp_3",
    "..."
  ],
  "videoUrl": null,
  "authorNickname": "逃喵喵",
  "authorId": "60598115000000000100ab29",
  "detailLevel": "full"
}

And a profile row from the same run:

{
  "type": "profile",
  "id": "5c32c9bb000000000502cb54",
  "nickname": "圆了个圆",
  "redId": "lxy17_",
  "bio": "我想我会一直在路上",
  "ipLocation": "江西",
  "tags": ["金牛座", "江西南昌", "旅行博主", "探店博主"],
  "followers": 10000,
  "followersText": "1万+",
  "notesPosted": 389,
  "notesLikedTotal": 317205,
  "notesCollectedTotal": 179950,
  "firstPageNotes": 32
}

Search by keyword

Xiaohongshu's own search page needs an account. This Actor never logs in, so it finds the notes for each keyword through public web search engines (notes Xiaohongshu itself exposes to them), then opens each note in full: text, tags, likes, saves, comments, shares, date, images, video link and author. Each row carries the searchKeyword it came from and the engine that found it (discoveredVia).

{
  "searchKeywords": ["露营", "露营装备"],
  "maxNotesPerKeyword": 50,
  "expandKeywords": true
}

• Use Chinese terms. Search engines index Xiaohongshu notes by their Chinese text: 露营 (camping) returned 40 notes on the first two result pages in our tests, 护肤 (skincare) 83 with variations, while skincare routine returned 1.
• These are indexed notes, not the newest posts. Search engines pick up a note weeks or months after it goes live, so most results are a few months old (in our 2026-10-17 tests the newest were from July 2026). There is no "latest" sort and no date filter. For fresh posts use the home feed (recommend).
• expandKeywords (default on) adds a few variations of each keyword (the keyword + 推荐, 攻略, 分享, 教程) when the exact keyword runs out, which adds notes the exact term did not find. With variations, a Chinese keyword usually reaches 80 to 150 notes.
• Every delivered note is charged as a full note (US$ 6.00 per 1,000). A note that does not open is skipped and not charged; the same note found by two keywords is delivered and charged once.

What you can do with it

• Influencer vetting (KOL / KOC): notes posted, total likes and saves, tags and IP location of a creator, plus their latest notes and how each one performed.
• Campaign tracking: paste the links of the notes of a campaign and get likes, saves, comments and shares on a schedule.
• Content and trend research on Chinese consumers: what the home feed recommends today, and what notes exist on a topic (露营, 护肤, 上海美食), with full text and hashtags.
• Media archiving: every image link and the MP4 video link of each note.

Input

Field	What it does
noteUrls	Note links copied from the browser: https://www.xiaohongshu.com/explore/<id>?xsec_token=.... A link without the token, or a bare note id, is tried too (notes indexed by search engines open without it). Short xhslink.com links are tried; one that does not resolve to a note comes back as a free NOTE_UNAVAILABLE row
searchKeywords	Keywords, best in Chinese (露营, 护肤). Notes found through search engines, each opened in full. Aliases: keywords, searchQueries, query
maxNotesPerKeyword	Default 30: full notes per keyword. Alias: maxResults
expandKeywords	Default on: also searches a few variations of each keyword (+ 推荐, 攻略, 分享, 教程) to reach the limit
profileUrls	https://www.xiaohongshu.com/user/profile/<id> or the 24-character user id
includeProfileNotes	Default on: one row per note on the profile's first page
profileNotesOrder	newest (default) sorts the profile's notes by publish date; site keeps the profile order, where pinned notes come first even when old. Every row has isPinned
skipPinnedNotes	Leave out the notes pinned to the top of the profile
profileNotesMaxAgeDays	Only notes published in the last N days (7 for a weekly run); empty keeps all
feedChannels	recommend (home feed), fashion, food, cosmetics, movies, career, love, home, gaming, travel, fitness (Chinese names also work)
maxItemsPerChannel	Default 30; each page brings about 30 notes
openFeedNotes	Opens each feed note for the full details (charged as a full note)
onlyNew	For scheduled runs: delivers, and charges, only notes not delivered in earlier runs (per profile, channel and keyword). Alias: onlyNewNotes

{
  "noteUrls": ["https://www.xiaohongshu.com/explore/6455b3d7000000000800e27f?xsec_token=ABj6wOII3wKf1doBy-UiXlhHC-HP-A9ammCgjhvhiT1L8%3D&xsec_source=pc_feed"],
  "profileUrls": ["5c32c9bb000000000502cb54"],
  "feedChannels": ["recommend"],
  "maxItemsPerChannel": 60,
  "openFeedNotes": true
}

Schedule it daily: only new notes

Turn on onlyNew, save the input as a Task and attach a daily Schedule:

{ "profileUrls": ["5c32c9bb000000000502cb54"], "feedChannels": ["recommend"], "maxItemsPerChannel": 30, "onlyNew": true }

• The first run delivers the profile with its first-page notes and the feed up to your limit, and remembers what it read.
• Every run after that delivers, and charges, only the notes that were not delivered before. The profile row comes only together with a new note, so a quiet profile costs nothing; the run ends green with the status "No new notes since the last run".
• The home feed is a fresh recommendation on every page, so onlyNew there means "never the same note twice", not "posted since yesterday".
• It works for keywords too: a scheduled keyword run delivers only the indexed notes it had not delivered before.
• The memory lives in a Key-Value Store named xiaohongshu-scraper-only-new in your own Apify account, one record per profile, channel and keyword. Delete it to start over. Note links are always returned.

Output fields

Notes: id, url (ready to open, with its token when Xiaohongshu gives one), xsecToken, noteType (normal = images, video), title, desc, tags, likedCount, collectedCount, commentCount, shareCount, publishedAt, lastUpdatedAt, ipLocation (when Xiaohongshu shows it), imageCount, images, coverUrl, videoUrl (MP4), videoDurationSec, authorNickname, authorId, authorUrl, isPinned (profile notes), detailLevel (full or card), searchKeyword and discoveredVia (brave, duckduckgo or bing; keyword search only, null otherwise), source, scrapedAt.

Note cards (feed or profile, not opened) carry title, noteType, likedCount, coverUrl, author and, from the feed, id and url; the other counters are null.

Profiles: id, url, nickname, redId, bio, ipLocation, gender, tags, followers, following, likesAndCollects (with the ...Text value exactly as shown, like 1万+), notesPosted, notesLikedTotal, notesCollectedTotal (exact), firstPageNotes.

Only what Xiaohongshu shows publicly about an author: nickname and public id. Never photos. No comments are collected.

Error codes (never charged)

errorCode	Meaning
XSEC_TOKEN_REQUIRED	The link has no xsec_token (or is a bare id) and the note did not open without it. Only notes indexed by search engines open without a token; copy the link from the browser with its token
NOTE_UNAVAILABLE	Deleted or private note, expired token, or an xhslink.com short link that did not resolve to a note
PROFILE_NOT_FOUND	No public profile with this id
NO_RESULTS	Nothing new for this input (duplicate or empty), or no indexed notes for a keyword (try a Chinese term)
INVALID_URL, INVALID_INPUT	A value the Actor does not understand
BLOCKED	Xiaohongshu or the search engines did not answer after retries; run again
NOT_REACHED, UNEXPECTED	The run timeout arrived, or something unforeseen broke; only that input is lost

Pricing

Event	Per 1,000
Full note delivered	US$ 6.00
Note card delivered	US$ 3.50
Profile delivered	US$ 20.00

Duplicates and error rows are free, and there is no Apify compute bill on top.

FAQ

How does keyword search work without an account? Xiaohongshu's search page is blank for a visitor without an account, but Xiaohongshu exposes many notes to web search engines, and those notes open for a logged-out visitor. The Actor finds those notes for your keyword and opens each one. So you get notes indexed by search engines (usually a few months old), not the live in-app ranking, and Chinese keywords work far better than English ones.

Why no comments? The comment list only loads for logged-in users. This Actor never logs in and never uses cookies, so it reads only what a logged-out visitor can see.

What is xsec_token and where do I get it? It is a signed token Xiaohongshu adds to every note link it hands out; the note page does not open without it. Open the note in a browser and copy the address bar: https://www.xiaohongshu.com/explore/<id>?xsec_token=...&xsec_source=.... Notes from the home feed and opened feed notes already come with their token in the url and xsecToken fields, so you can feed them back as noteUrls.

Is this a Xiaohongshu API alternative? Yes, for public data. Xiaohongshu has no public API; this Actor returns notes, profiles and feeds as JSON, CSV or Excel through the Apify API, without login, cookies or app keys. Anything that needs an account (the in-app search ranking, comments, private notes) is out of scope.

Is RedNote the same as Xiaohongshu / Little Red Book? Yes. Xiaohongshu (小红书) is the Chinese name, Little Red Book its literal translation, and RedNote the English name the app uses internationally. This Actor reads the same site, xiaohongshu.com.

Limitations

• No comments, and keyword search covers indexed notes only: the in-app search and comments need a logged-in account, and this Actor never logs in. Keyword results come from web search engines: mostly notes a few months old, no newest-first sort, few results for English terms.
• A fresh note needs its xsec_token. Links copied from the website carry it; notes indexed by search engines also open without it. xhslink.com short links from the app are tried, with no promise: if one does not resolve to a note, or the token expired, the row comes back as a free NOTE_UNAVAILABLE.
• Profile notes have no link. For logged-out visitors Xiaohongshu blanks the note ids on profile pages, so those rows carry title, type, likes, date and cover, but cannot be opened. Only the first page (about 30 notes) is shown without login.
• Topic channels serve older notes. To logged-out visitors the topic channels (food, travel...) come from an older selection (notes from 2023, measured on 2026-09-30). The home feed (recommend) is current and brings new notes on every page.
• Image and video links are signed by Xiaohongshu and expire after some time; download them soon after the run.
• Only public content. This Actor reads public pages only and respects the site's terms.

Related scrapers

• WeChat Official Account Scraper: WeChat articles with full text, by keyword, account or link.
• Weibo Scraper: Weibo posts, comments, profiles and hot search.
• Douyin Hot Search Scraper: China trends from Douyin, Baidu, Zhihu and Toutiao.
• Bilibili Scraper: Bilibili videos, comments, channels and stats.
• Douban Scraper: Douban reviews and ratings of movies and books.

Changelog

• 2026-10-17: keyword search (searchKeywords, maxNotesPerKeyword, expandKeywords): notes found through web search engines, each opened in full, with searchKeyword and discoveredVia. Same price as a note by link. Note links without xsec_token (and bare ids) are now tried before being refused.
• 2026-10-04: onlyNew (alias onlyNewNotes) for scheduled runs: only notes not delivered before are returned and charged; the profile row comes only with a new note. Same price.
• 2026-09-30, 0.1: first release. Notes by link (web links with their token; xhslink.com short links tried), profiles with first-page notes, home feed and topic channels, optional full details for feed notes, free error rows.