RedNote (Xiaohongshu) Search Scraper

Search 小红书 notes by keyword and get structured rows: title, author, engagement counts, cover, and a ready-to-scrape note URL with its xsec_token.

Xiaohongshu (RedNote / Little Red Book) gates keyword search behind login. This actor signs every request through Xiaohongshu's own web signing (x-s / x-t) using a headless page, so your searches behave like a real browser session — you just supply a web_session cookie from an account you control.

What you get per note

Field	Description
noteId	24-hex note ID
noteType	normal (image post) or video
noteTitle	Display title
likedCount / likedCountNum	Likes as shown (e.g. 1.2万) and parsed to a number (12000)
collectedCount, commentCount	Saves and comments
userId, userName, userAvatar	Author
noteCoverUrl	Cover image URL
noteUrl + xsecToken	Ready to paste into the Note Detail or Comments scraper

Sort by comprehensive (Xiaohongshu's default ranking), most popular, or newest. Paginate up to 100 pages (~20 notes per page), cap total rows with maxItems.

How to get the web_session cookie

1. Log in at xiaohongshu.com in your browser.
2. Open DevTools → Application → Cookies → https://www.xiaohongshu.com.
3. Copy the value of web_session and paste it into the input. Pasting your whole cookie string also works — the actor extracts what it needs.

The cookie stays yours: it is marked secret in the input schema and is never logged.

Example input

{
    "keyword": "skincare",
    "searchSort": "popularity_descending",
    "maxPages": 3,
    "webSession": "<your web_session value>"
}

FAQ

Why do I need a cookie at all? Xiaohongshu returns a login wall for guest search API calls. Note detail, user profiles, and category feeds are open — search and comments are not. If you only need a single note's data, the Note Detail scraper needs no login.

The run fails immediately with a session error. The web_session value is expired or from a logged-out browser. Log in again and copy a fresh value — the actor fails fast instead of burning your budget on doomed requests.

Are the Chinese counts usable in spreadsheets? Yes — every count comes twice: as displayed (1.2万) and parsed (12000) in the *Num twin field.

Related scrapers

• RedNote Note Detail Scraper — one note's full data, no login
• RedNote Comments Scraper — full comment threads
• Xiaohongshu (RedNote) Scraper — the all-in-one actor: 8 operations including user profiles, channel feeds, and mall products

🤖 For AI Agents & LLM Apps

Compact reference for AI agents calling this actor via the Apify MCP server or the Apify API (actor: memo23/rednote-search-scraper).

Purpose: Keyword-searches Xiaohongshu (RedNote / 小红书) notes and returns structured rows — title, author, engagement counts, cover image, and a ready-to-scrape note URL with its xsec_token.

Minimal input:

{
  "keyword": "skincare",
  "webSession": "<your web_session cookie value>",
  "searchSort": "general",
  "maxPages": 1,
  "maxItems": 20
}

Output: one dataset row per note; key fields: noteId, noteType, noteTitle, likedCount, likedCountNum, collectedCount, commentCount, userId, userName, userAvatar, noteCoverUrl, noteUrl, xsecToken. Each engagement count also has a numeric *Num twin (Chinese 1.2万 → 12000).

Behaviors an agent should know:

• Requires a logged-in Xiaohongshu web_session cookie — keyword search is login-gated. An expired/logged-out value fails the run fast. webSession is marked secret and never logged.
• Always set maxItems (and/or maxPages, ~20 notes per page) to cap output; uncapped runs paginate up to 100 pages.
• searchSort enum: general (comprehensive, default), popularity_descending (most popular), time_descending (newest).
• The noteUrl + xsecToken in each row paste directly into the RedNote Note Detail or Comments scraper for fan-out.
• Billing: Pay-per-event billing — see the Pricing tab on the actor page.

⚠️ Disclaimer

This actor collects publicly displayed data from Xiaohongshu for legitimate research, marketing, and archival purposes. You are responsible for complying with applicable laws, Xiaohongshu's terms, and for the account whose cookie you supply. Do not use collected data for spam or harassment.