---
title: "蒲公英内容广场（热门内容榜单）/PGY content square leaderboards"
method: POST
path: "/api/v1/xiaohongshu/pgy/get_content_square"
tags: ["Xiaohongshu-PGY-API"]
---

# 蒲公英内容广场（热门内容榜单）/PGY content square leaderboards

`POST /api/v1/xiaohongshu/pgy/get_content_square`

## Request

- Base URL: `https://api.tikhub.io` (the document declares no server; this is the origin it was published from)
- URL: `https://api.tikhub.io/api/v1/xiaohongshu/pgy/get_content_square`
- Auth: HTTP bearer

# [中文]
### 用途:
- 复刻蒲公英「学内容 → 创意灵感 → 内容广场」：**不需要任何 ID**，直接查「现在什么内容在火」
- 与其它接口方向相反 —— 其余接口都要先有博主/笔记 ID，本接口是入口式的，
  拿到榜单后可以顺着笔记里的博主 ID 再去调博主侧接口深挖

### 1. 全部参数一览
| 参数 | 必填 | 默认 | 说明 |
|---|---|---|---|
| `biz_type` | ✅ | — | 榜单，见第 3 节 |
| `page_num` | | `1` | 页码，1-100 |
| `page_size` | | `34` | 每页数量，**1-34**（网页固定 34） |
| `search_word` | | `""` | 关键词；留空=看整榜。支持英文逗号分隔批量搜 |
| `order_by` | | 随榜单 | 排序字段，见第 5 节；留空=用该榜默认排序 |
| `date_range` | | `7` | 统计窗口，取 `3` / `7` / `14` / `30`（近 N 日） |
| `sub_biz_type` | | `3` | **仅榜 8**，见第 6 节 |
| `category` | | — | **仅榜 1** 的类目，见第 4 节 |
| `industry` | | — | **榜 2/3/4/5/6/7/8** 的所属行业，见第 4 节 |
| `note_type` / `content_type` / `theme` | | — | 见第 6 节 |
| `author_type` / `placement` / `marketing_target` | | — | 见第 6 节 |
| `price` / `fans` | | — | **仅榜 4**，见第 6 节 |

不传的筛选就是「不限」。**不要把不需要的字段传成空字符串或 0** —— 那会被当成一次真实筛选。

### 2. 翻页
翻页只改 `page_num`，其余参数**必须原样保持不变**。

**每个榜单固定只有 100 条**（返回的 `total` 恒为 100）—— 这是榜单不是全量搜索，
上游只放出 Top 100。所以 `page_size=34` 时翻到第 3 页就到底了，第 4 页返回空数组。
要更多内容请换榜单、换筛选或换 `date_range`，**调大 `page_size` 没用**（上限 34）。

响应里的 `pageInfoDto.totalPage` 实测恒为 `0`，**不要拿它判断还有没有下一页**。

### 3. 榜单（`biz_type`）
| 值 | 榜单 | 内容 |
|---|---|---|
| `1` | 小红书热门 | 全平台热门内容榜单 |
| `2` | 产品种草 | 营销诉求为产品种草的内容榜单 |
| `3` | 电商推广 | 营销诉求为商品或直播推广的内容榜单 |
| `4` | 蒲公英合作 | 笔记博主合作的内容榜单 |
| `5` | 客资收集 | 营销诉求为客资收集的内容榜单 |
| `6` | 电商热门 | 电商热门商品内容榜单 |
| `7` | 种草直达 | 营销诉求为种草直达的内容榜单 |
| `8` | 应用推广 | 营销诉求为应用推广的内容榜单 |

### 4. 各榜支持的筛选不同
榜单决定了可用的筛选项 —— 传了该榜不支持的字段会返回 **422** 并告诉你这个榜支持哪些，
不会被静默忽略（否则你会拿到一份没按预期过滤的数据却发现不了）。

| 榜 | 可用筛选 |
|---|---|
| `1` 小红书热门 | `category`、`note_type`、`content_type` |
| `2` 产品种草 / `5` 客资收集 / `7` 种草直达 | `industry`、`placement`、`note_type`、`content_type`、`theme` |
| `3` 电商推广 | `industry`、`placement`、`note_type`、`content_type`、`marketing_target` |
| `4` 蒲公英合作 | `industry`、`note_type`、`author_type`、`theme`、`price`、`fans` |
| `6` 电商热门 | `industry`、`note_type`、`content_type` |
| `8` 应用推广 | `industry`、`placement`、`note_type`、`content_type`、`theme`、`sub_biz_type` |

**行业怎么填**（不同榜是不同的分类体系，别混用）：

| 榜 | 字段 | 取值 |
|---|---|---|
| `1` | `category` | `内容类目#X` 或 `所属行业#X`。内容类目 X 取：美妆、护肤、个人护理、母婴、时尚、美食、家居家装、影视综资讯、运动健身、宠物、文化艺术、兴趣爱好、生活记录、教育、职场、情感、摄影、游戏、科技数码、出行旅游、音乐、搞笑、健康养生、汽车、婚嫁、商业财经、素材、其他 |
| `2/4/5/7/8` | `industry` | 母婴、家用电器、3C数码、食品饮料、美妆个护、汽车出行、本地生活、日化家清、医疗健康、宠物（新）、家居家装、出行旅游、教育培训、金融行业、医疗医美、免税平台、服饰鞋包、珠宝配饰、运动户外、文玩娱乐、到店综合、奢侈品、互联网、影像婚美、房地产 |
| `3/6` | `industry` | 电商商品类目，如 茶、彩妆/香水/美妆工具、厨房电器、床上用品……（数百项，与上面两套完全不同） |

### 5. 排序（`order_by`）
留空即用该榜的网页默认排序（榜 2 和 3 默认阅读率，其余默认曝光量）。

通用（所有榜可用）：

| 取值 | 含义 | 取值 | 含义 |
|---|---|---|---|
| `premium_imp_num` | 曝光量 | `premium_engage_rate` | 互动率 |
| `premium_read_num` | 阅读量 | `premium_like_num` | 点赞量 |
| `premium_ctr` | 阅读率 | `premium_fav_num` | 收藏量 |
| `premium_engage_num` | 互动量 | `premium_cmt_num` | 评论量 |

仅 `3` 电商推广 / `6` 电商热门 额外支持：`premium_ecm_dgmv`（笔记支付金额）、
`premium_ecm_order_pv`（笔记支付订单数）、`premium_ecm_order_rate`（笔记支付转化率）。

仅 `4` 蒲公英合作 额外支持（小红星=淘宝、小红盟=京东、小红链=唯品会）：
`taobao_third_active_pgy_user_ratio` / `taobao_third_active_pgy_user_cost` /
`jd_third_active_pgy_user_ratio` / `jd_third_active_pgy_user_cost` /
`wph_third_active_pgy_user_ratio` / `wph_third_active_pgy_user_cost`。

传了本榜不支持的排序同样返回 422。

### 6. 其它枚举一览
| 参数 | 网页名称 | 全部取值 |
|---|---|---|
| `note_type` | 内容类型 | `1` 图文、`2` 视频 |
| `content_type` | 内容来源 | `1` 博主合作笔记、`2` 明星合作笔记、`3` 员工笔记、`5` 买手笔记、`6` 专业号笔记、`11` 主理人笔记、`-1` 用户笔记 |
| `author_type` | 创作者类型 | `1` 蒲公英博主、`2` 明星 |
| `placement` | 流量场域 | `1` 发现、`2` 搜索、`3` 视频内流 |
| `marketing_target` | 营销目标 | `3` 商品推广、`8` 直播推广 |
| `sub_biz_type` | 应用唤起 | `1` 应用下载、`2` 小程序推广、`3` 应用唤起 |
| `date_range` | 时间范围 | `3`、`7`、`14`、`30`（近 N 日） |
| `theme` | 题材 | 吃播、双品对比、活动分享、选购攻略、问答类、仿妆变装、采访类、单品推荐、经验教程、探店打卡、生活剧情、新品预告、创意内容、知识科普、测评类、好物合集、试色、情感话题、OOTD、手工制作、开箱 |

`price`（创作者报价，仅榜 4）要同时给口径和区间，因为图文和视频是两套独立报价：
`{"kind": 1, "range": {"min": 10000, "max": 50000}}`，`kind` 取 `1` 图文笔记 / `2` 视频笔记。
`fans`（创作者粉丝，仅榜 4）只要区间：`{"min": 100000, "max": 500000}`。

### 7. 返回
- `noteList[]`: 命中的笔记。每条含 `noteInfo`（`title` 标题、`noteType` 类型、
  `noteLink` 笔记链接、`noteImages` 封面、`readNum` 阅读、`likeNum` 点赞、`favNum` 收藏、
  `cmtNum` 评论、`videoDuration` 视频时长、`notePublishTime` 发布时间）与博主信息
- `total`: 该榜的条数，恒为 **100**
- `pageInfoDto`: 上游分页回执，其中 `totalPage` 恒为 `0`，**不可用**

### 8. 注意:
- **字段名写错、枚举值不合法、榜单与筛选不匹配都会 422**
- 关键词支持英文逗号分隔批量搜（与网页搜索框一致）

### 响应结构（取数要剥两层）:
- `resp["data"]` 是蒲公英原样信封（含 `code`/`msg`/`success`），业务数据还在它里面一层
- 正确取法：`data = resp["data"]["data"]`；下面「返回」列的字段都在这一层下
### 计费:
- HTTP 200 → **计费**。包含「查无结果」（`data` 为 null，如 ID 不存在或筛选无命中）——
  请求已送达蒲公英，同样扣费，不要因为拿不到数据就重试
- HTTP 400 → 不计费（参数格式非法，或上游资源不足/异常）

# [English]
### Purpose:
- Replica of PGY "Learn Content → Inspiration → Content Square": **no ID required**,
  directly answers "what content is hot right now"
- Opposite direction from every other endpoint — this one is an entry point; take the
  blogger IDs from the returned notes and drill down with the blogger endpoints

### 1. Full Parameter List
| Param | Required | Default | Description |
|---|---|---|---|
| `biz_type` | ✅ | — | Leaderboard, see section 3 |
| `page_num` | | `1` | Page, 1-100 |
| `page_size` | | `34` | **1-34** (fixed at 34 on the web page) |
| `search_word` | | `""` | Keyword; empty = whole board. Comma-separated for batch search |
| `order_by` | | per board | Sort field, see section 5; empty = board default |
| `date_range` | | `7` | Window in days: `3` / `7` / `14` / `30` |
| `sub_biz_type` | | `3` | **Board 8 only**, see section 6 |
| `category` | | — | **Board 1 only**, see section 4 |
| `industry` | | — | **Boards 2/3/4/5/6/7/8**, see section 4 |
| `note_type` / `content_type` / `theme` | | — | See section 6 |
| `author_type` / `placement` / `marketing_target` | | — | See section 6 |
| `price` / `fans` | | — | **Board 4 only**, see section 6 |

Omitted filters mean "no limit". **Do not send unused fields as empty strings or 0** —
they would be treated as a real filter.

### 2. Paging
Change `page_num` only and keep every other parameter identical.

**Each board holds exactly 100 items** (`total` is always 100) — this is a leaderboard,
not a full search; upstream only exposes the Top 100. With `page_size=34` you reach the
end at page 3, and page 4 returns an empty array. For more content switch boards, filters
or `date_range` — **raising `page_size` does not help** (capped at 34).

`pageInfoDto.totalPage` is always `0` in practice — **do not use it** to detect more pages.

### 3. Leaderboards (`biz_type`)
| Value | Board | Content |
|---|---|---|
| `1` | Xiaohongshu Hot | Platform-wide hot content |
| `2` | Product Seeding | Content aimed at product seeding |
| `3` | E-commerce Promotion | Content promoting goods or live streams |
| `4` | PGY Cooperation | Sponsored notes from PGY bloggers |
| `5` | Lead Collection | Content aimed at lead collection |
| `6` | E-commerce Hot | Hot e-commerce product content |
| `7` | Direct Seeding | Content aimed at direct seeding |
| `8` | App Promotion | Content promoting apps |

### 4. Filters Differ per Board
The board determines which filters are available — sending an unsupported field returns
**422** listing what the board does support, instead of being silently ignored (otherwise
you would receive unfiltered data without noticing).

| Board | Available filters |
|---|---|
| `1` | `category`, `note_type`, `content_type` |
| `2` / `5` / `7` | `industry`, `placement`, `note_type`, `content_type`, `theme` |
| `3` | `industry`, `placement`, `note_type`, `content_type`, `marketing_target` |
| `4` | `industry`, `note_type`, `author_type`, `theme`, `price`, `fans` |
| `6` | `industry`, `note_type`, `content_type` |
| `8` | `industry`, `placement`, `note_type`, `content_type`, `theme`, `sub_biz_type` |

**How to fill industry** (different boards use different taxonomies — do not mix them):

| Board | Field | Values |
|---|---|---|
| `1` | `category` | `内容类目#X` or `所属行业#X`, e.g. `内容类目#美妆` |
| `2/4/5/7/8` | `industry` | Industry names (母婴, 美妆个护, 3C数码, ... 25 in total) |
| `3/6` | `industry` | E-commerce product categories (茶, 厨房电器, ... hundreds, a different taxonomy) |

### 5. Sorting (`order_by`)
Leave empty to use the board default (boards 2 and 3 default to read rate, others to impressions).

Common to all boards:

| Value | Meaning | Value | Meaning |
|---|---|---|---|
| `premium_imp_num` | Impressions | `premium_engage_rate` | Engagement rate |
| `premium_read_num` | Reads | `premium_like_num` | Likes |
| `premium_ctr` | Read rate | `premium_fav_num` | Favorites |
| `premium_engage_num` | Engagements | `premium_cmt_num` | Comments |

Boards `3` / `6` additionally support `premium_ecm_dgmv` (paid GMV),
`premium_ecm_order_pv` (paid orders), `premium_ecm_order_rate` (paid conversion rate).

Board `4` additionally supports off-site metrics (小红星=Taobao, 小红盟=JD, 小红链=VIP):
`taobao_third_active_pgy_user_ratio` / `_cost`, `jd_third_active_pgy_user_ratio` / `_cost`,
`wph_third_active_pgy_user_ratio` / `_cost`.

Sending a sort value the board does not support also returns 422.

### 6. Other Enums
| Param | Web name | All values |
|---|---|---|
| `note_type` | Content type | `1` image-text, `2` video |
| `content_type` | Content source | `1` blogger sponsored, `2` celebrity sponsored, `3` employee, `5` buyer, `6` pro account, `11` owner, `-1` user (values are non-contiguous) |
| `author_type` | Creator type | `1` PGY blogger, `2` celebrity |
| `placement` | Traffic placement | `1` discovery, `2` search, `3` in-video feed |
| `marketing_target` | Marketing target | `3` product promotion, `8` live promotion |
| `sub_biz_type` | App action | `1` app download, `2` mini-program, `3` app open |
| `date_range` | Time window | `3`, `7`, `14`, `30` (last N days) |
| `theme` | Theme | 21 Chinese theme names (开箱, 测评类, 好物合集, ...) |

`price` (board 4) needs both the quote type and the range, because image-text and video are
two independent quotes: `{"kind": 1, "range": {"min": 10000, "max": 50000}}` where `kind`
is `1` image-text / `2` video. `fans` (board 4) takes a plain range.

### 7. Return
- `noteList[]`: matched notes; each has `noteInfo` (title, noteType, noteLink, noteImages,
  readNum, likeNum, favNum, cmtNum, videoDuration, notePublishTime) plus blogger info
- `total`: always **100** (board capacity)
- `pageInfoDto`: upstream paging echo; its `totalPage` is always `0` and unusable

### 8. Notes:
- **Misspelled fields, invalid enums, and board/filter mismatches all return 422**
- Keyword supports comma-separated batch search (same as the web search box)

### Response shape (two layers):
- `resp["data"]` is PGY's raw envelope (with `code`/`msg`/`success`); the payload is one level deeper
- Correct access: `data = resp["data"]["data"]` — every field listed under "Return" lives here
### Billing:
- HTTP 200 → **charged**, including "no result" (`data` is null, e.g. ID not found or filters matched
  nothing): the request did reach PGY, so retrying on empty data just burns quota
- HTTP 400 → not charged (malformed params, or upstream unavailable/error)

# [示例/Example]
```json
// ① 近 7 日全平台最热的美妆视频（榜 1 用 category，不是 industry）
{"biz_type": 1, "category": "内容类目#美妆", "note_type": 2}

// ② 母婴行业近 30 日互动率最高的蒲公英合作笔记，博主图文报价 1~5 万
{"biz_type": 4, "industry": "母婴", "date_range": 30,
 "order_by": "premium_engage_rate",
 "price": {"kind": 1, "range": {"min": 10000, "max": 50000}}}

// ③ 电商推广榜按笔记支付金额排序，只看搜索场域
{"biz_type": 3, "order_by": "premium_ecm_dgmv", "placement": 2}

// ④ 带关键词看榜
{"biz_type": 1, "search_word": "面膜"}

// ⑤ 批量关键词（英文逗号分隔）+ 只看博主合作笔记
{"biz_type": 1, "search_word": "面膜,精华", "content_type": 1}

// ⑥ 翻页：只改 page_num，其余参数原样不动（改了口径两页会重复/遗漏）；每榜只有 100 条，34/页翻到第 3 页到底
{"biz_type": 1, "page_num": 1, "page_size": 34, "category": "内容类目#美妆"}
{"biz_type": 1, "page_num": 2, "page_size": 34, "category": "内容类目#美妆"}

// ⑦ 蒲公英合作榜（仅榜 4 支持站外指标）：按小红星站外转化率排，限蒲公英博主、10~50 万粉的视频
{"biz_type": 4, "order_by": "taobao_third_active_pgy_user_ratio", "author_type": 1,
 "note_type": 2, "fans": {"min": 100000, "max": 500000}}

// ⑧ 产品种草榜近 3 日：美妆个护行业、测评题材、发现页场域的图文（留空 order_by = 用该榜默认排序）
{"biz_type": 2, "industry": "美妆个护", "theme": "测评类", "placement": 1,
 "note_type": 1, "date_range": 3}

// ⑨ 电商推广榜只看直播推广（marketing_target 仅榜 3 可用），按支付订单数排
{"biz_type": 3, "marketing_target": 8, "order_by": "premium_ecm_order_pv", "note_type": 2}

// ⑩ 电商热门榜近 14 日：industry 这里是电商商品类目（与榜 2/4/5/7/8 的行业名不是一套），看买手笔记的支付转化率
{"biz_type": 6, "industry": "厨房电器", "order_by": "premium_ecm_order_rate",
 "content_type": 5, "date_range": 14}

// ⑪ 应用推广榜（sub_biz_type 仅榜 8 可用）：应用下载口径、互联网行业、发现页的知识科普题材
{"biz_type": 8, "sub_biz_type": 1, "industry": "互联网", "placement": 1, "theme": "知识科普"}

// 取数：notes = resp["data"]["data"]["noteList"]
```

## Request body

- ContentSquareReq
  - `biz_type` 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8, required: 榜单 1=小红书热门 2=产品种草 3=电商推广 4=蒲公英合作 5=客资收集 6=电商热门 7=种草直达 8=应用推广/Leaderboard
  - `page_num` integer: 页码；每个榜固定只有 100 条/Page; each board has only 100 items
  - `page_size` integer: 每页数量，1-34（网页固定 34）/Page size, 1-34
  - `search_word` string: 关键词，留空=看整榜；支持英文逗号分隔批量搜/Keyword, comma-separated for batch
  - `order_by` string: 排序字段，留空=用该榜默认排序/Sort field, empty=board default
  - `date_range` 3 | 7 | 14 | 30: 统计窗口：近 3/7/14/30 日/Date window in days
  - `sub_biz_type` 1 | 2 | 3, nullable: 【仅榜 8】1=应用下载 2=小程序推广 3=应用唤起(默认)/Board 8 only
  - `category` string, nullable: 【仅榜 1】类目，格式「一级#二级」，如 "内容类目#美妆"/Board 1 only
  - `industry` string, nullable: 【榜 2/3/4/5/6/7/8】所属行业名，取值集因榜而异/Industry, values differ per board
  - `note_type` 1 | 2, nullable: 内容类型 1=图文 2=视频/Content type
  - `content_type` 1 | 2 | 3 | 5 | 6 | 11 | -1, nullable: 内容来源 1=博主合作 2=明星合作 3=员工 5=买手 6=专业号 11=主理人 -1=用户/Content source
  - `theme` string, nullable: 【榜 2/4/5/7/8】题材名/Theme
  - `author_type` 1 | 2, nullable: 【仅榜 4】1=蒲公英博主 2=明星/Board 4 only
  - `placement` 1 | 2 | 3, nullable: 【榜 2/3/5/7/8】流量场域 1=发现 2=搜索 3=视频内流/Traffic placement
  - `marketing_target` 3 | 8, nullable: 【仅榜 3】3=商品推广 8=直播推广/Board 3 only
  - `price` PriceFilter: 网页「创作者报价」：图文和视频是两个独立口径，必须先选一个。
    - `kind` 1 | 2, required: 报价口径：1=图文笔记 2=视频笔记
    - `range` CrawlersXiaohongshuPgyPgyContentFiltersRangeFilter, required: 数值区间。两端可省略其一，至少给一个。
      - `min` integer, nullable: 最小值（含）
      - `max` integer, nullable: 最大值（含）；省略=不限
  - `fans` CrawlersXiaohongshuPgyPgyContentFiltersRangeFilter: 数值区间。两端可省略其一，至少给一个。
    - `min` integer, nullable: 最小值（含）
    - `max` integer, nullable: 最大值（含）；省略=不限

## Response `200`

Successful Response

- ResponseModel
  - `code` integer: HTTP status code | HTTP状态码
  - `request_id` string, nullable: Unique request identifier | 唯一请求标识符
  - `message` string: Response message (EN-US) | 响应消息 (English)
  - `message_zh` string: Response message (ZH-CN) | 响应消息 (中文)
  - `support` string: Support message | 支持消息
  - `time` string: The time the response was generated | 生成响应的时间
  - `time_stamp` integer: The timestamp the response was generated | 生成响应的时间戳
  - `time_zone` string: The timezone of the response time | 响应时间的时区
  - `docs` string, nullable: Link to the API Swagger documentation for this endpoint | 此端点的 API Swagger 文档链接
  - `cache_message` string, nullable: Cache message (EN-US) | 缓存消息 (English)
  - `cache_message_zh` string, nullable: Cache message (ZH-CN) | 缓存消息 (中文)
  - `cache_url` string, nullable: The URL to access the cached result | 访问缓存结果的 URL
  - `router` string: The endpoint that generated this response | 生成此响应的端点
  - `params` unknown
  - `data` unknown

## Other responses

- `422`: Validation Error

## Changes

- **2026-10-08** `96a122a5f220`: 2 info
  - added the optional property `detail/items/ctx` to the response with the `422` status
  - added the optional property `detail/items/input` to the response with the `422` status
- **2026-10-06** `927ffb5aef5d`: 1 breaking, 1 info
  - removed `#/components/schemas/RangeFilter` from the `fans` request property `anyOf` list
  - added `#/components/schemas/crawlers__xiaohongshu__pgy__pgy_content_filters__RangeFilter` to the `fans` request property `anyOf` list
- **2026-10-05** `83a77e4ee893`: 1 breaking, 1 info
  - removed `#/components/schemas/crawlers__xiaohongshu__pgy__pgy_content_filters__RangeFilter` from the `fans` request property `anyOf` list
  - added `#/components/schemas/RangeFilter` to the `fans` request property `anyOf` list
- **2026-10-04** `29d100673358`: 1 breaking, 1 info
  - removed `#/components/schemas/RangeFilter` from the `fans` request property `anyOf` list
  - added `#/components/schemas/crawlers__xiaohongshu__pgy__pgy_content_filters__RangeFilter` to the `fans` request property `anyOf` list
- **2026-10-02** `428b27c5eeb0`: 1 breaking, 1 info
  - removed `#/components/schemas/crawlers__xiaohongshu__pgy__pgy_content_filters__RangeFilter` from the `fans` request property `anyOf` list
  - added `#/components/schemas/RangeFilter` to the `fans` request property `anyOf` list

[Operation history](https://skmtc.dev/tikhub/apis/tikhub-douyin-tiktok-xiaohongshu-lemon8-bilibili-kuaishou-pi/changes/api/v1/xiaohongshu/pgy/get_content_square/post.md)

---

[API](https://skmtc.dev/tikhub/apis/tikhub-douyin-tiktok-xiaohongshu-lemon8-bilibili-kuaishou-pi.md) - [All operations](https://skmtc.dev/tikhub/apis/tikhub-douyin-tiktok-xiaohongshu-lemon8-bilibili-kuaishou-pi/llms.txt) - [OpenAPI document](https://skmtc.dev/tikhub/apis/tikhub-douyin-tiktok-xiaohongshu-lemon8-bilibili-kuaishou-pi/revisions/86a3483c705a?raw)