# Do I need an xsec_token to get a Xiaohongshu note, and how do I get one from a note ID?

> Why Xiaohongshu note links and web API calls now need xsec_token, why it cannot be computed from a note ID, and how to fetch a note by ID or link without handling tokens.

Canonical page: https://okdata.online/answers/xiaohongshu-xsec-token/

**Short answer: **On Xiaohongshu's own web API, yes: a note opened without the xsec_token that came with it from a search or feed is often refused, and the token cannot be derived from the note ID. Through okdata you pass just the note ID or a share link and get the note back for $0.02; tokens are handled on our side.

You have a spreadsheet of note IDs from last month's campaign and want fresh like and save counts. Opening `xiaohongshu.com/explore/<id>` in a script returns an empty page, a "note unavailable" message or a 461, while the same link copied from the app works. The difference is the `xsec_token` query parameter. Xiaohongshu attaches it to every note it shows you in search results, feeds and profiles, and checks it when the note is opened. It is issued by the server, tied to where you found the note, and expires. There is no formula to turn a note ID into a valid token, which is why "how do I get xsec_token from note_id" keeps coming up in scraper issue trackers.

## Fetch by ID or link instead

[Xiaohongshu note detail](/apis/xiaohongshu/notes/get/) takes either `id` (the 24-character note ID) or `url` (a note page link or an xhslink.com share link). You do not send a token. You get the title, full text, every image or the video, likes, saves, comment and share counts, hashtags and the publish time.

If you are starting from keywords, [note search](/apis/xiaohongshu/notes/search/) returns note IDs ready for the detail call.

## What it costs

| Notes to refresh | Calls | Cost |
|---|---|---|
| 1 | 1 | $0.02 |
| 100 | 100 | $2.00 |
| 1,000 | 1,000 | $20.00 |

If all you need are the counts of notes that rank for a keyword, search results already carry likes, saves and comment counts for about 20 notes per call.

## Example

```python
import os, requests

ids = ["6650a1b2000000001e00abcd"]
for note_id in ids:
    r = requests.get(
        "https://api.okdata.online/v1/xiaohongshu/notes/get",
        headers={"Authorization": f"Bearer {os.environ['OKDATA_API_KEY']}"},
        params={"id": note_id, "fields": "id,title,metrics.likes,metrics.favorites,metrics.comments"},
        timeout=60,
    )
    print(r.json()["data"]["item"])
```

With `view=full`, the token Xiaohongshu attached to the note, when there is one, comes back under `ext.xsec_token`, handy if you want to build a link a person can open.

## Gotchas

- Deleted or private notes return `not_found`, which is not charged.
- `metrics.views` is null: Xiaohongshu does not publish view counts.
- There is no batch call for note detail; each note is one call and one charge.

## Alternatives

Open-source tools such as XHS-Downloader and Spider_XHS take a full link with its token and work well for one-off downloads from your own browser session. They break when the token is missing or stale. If you have only IDs, or you refresh the same notes every week, the ID-based call is simpler. The [competitor monitoring](/solutions/competitor-monitoring/) page shows a weekly refresh.

## Endpoints used

- [Xiaohongshu note detail](https://okdata.online/apis/xiaohongshu/notes/get/) `xiaohongshu.notes.get`: $0.02 per call
- [Xiaohongshu note search](https://okdata.online/apis/xiaohongshu/notes/search/) `xiaohongshu.notes.search`: $0.02 per call · about 20 notes per page

Updated 2026-11-12