Xiaohongshu (RedNote) user post keyword monitoring and WeChat notification. Use when the user needs to (1) monitor a specific Xiaohongshu user's latest posts...
---
name: xiaohongshu-trader-monitor
description: Xiaohongshu (RedNote) user post keyword monitoring and WeChat notification. Use when the user needs to (1) monitor a specific Xiaohongshu user's latest posts for keywords in titles, (2) get notified via WeChat when target keywords appear in new posts, (3) set up automated periodic checks for a Xiaohongshu trader/seller's updates, or (4) configure keyword-based alerts for Xiaohongshu content. Triggers on phrases like "小红书监测", "远行商人", "帖子监控", "关键词提醒", "xiaohongshu monitor".
---
# Xiaohongshu Trader Monitor
监测指定小红书用户的最新帖子,标题含关键词时自动推送到微信。
## ⚠️ 使用前配置(必须)
本 skill 需要你填入自己的认证信息才能运行。打开 `scripts/xhs_trader_monitor.py`,修改**用户配置区**(文件顶部):
### 1. 微信推送目标 ID
获取你的 openclaw-weixin 用户 ID:
```bash
openclaw message list-accounts --channel openclaw-weixin
```
将 `TARGET_ID` 替换为你的 ID,格式为 `xxx@im.wechat`。
### 2. 小红书目标用户信息
**USER_ID**:目标用户主页 URL 中的 ID
- 格式:`https://www.xiaohongshu.com/user/profile/USER_ID`
- 从分享链接或搜索结果中获取
**USER_XSEC**:xsec_token(认证令牌)
- 从目标用户任意帖子的 feed 数据中提取 `xsecToken`
- 或运行 `mcp-call.sh user_profile` 后从返回数据中获取
### 3. 监测关键词
在 `KEYWORDS` 列表中配置你要监测的关键词:
```python
KEYWORDS = ["棱镜球", "炫彩蛋", "同乘蛋", "祝福项坠"]
```
### 4. Cookies 登录
将已登录小红书的浏览器 cookies 导出为 JSON 数组,保存到 `cookies.json`:
```json
[
{"name": "a1", "value": "...", "domain": ".xiaohongshu.com", "path": "/"},
{"name": "web_session", "value": "...", "domain": ".xiaohongshu.com", "path": "/"}
```
放置路径(按优先级):
1. `~/cookies.json`
2. `~/.xiaohongshu/cookies.json`
3. 环境变量 `XHS_COOKIES_SRC` 指定的路径
> ⚠️ 首次登录可通过 `mcp-call.sh get_login_qrcode` 获取二维码扫码登录,但当前环境多被风控,推荐直接导出 cookies。
## Quick Start
### 安装依赖
```bash
# 1. 安装 xiaohongshu-mcp 二进制
# 从 GitHub Releases 下载对应平台文件:
# https://github.com/xpzouying/xiaohongshu-mcp/releases
mkdir -p ~/.local/bin
mv xiaohongshu-mcp-linux-amd64 ~/.local/bin/xiaohongshu-mcp
mv xiaohongshu-login-linux-amd64 ~/.local/bin/xiaohongshu-login
chmod +x ~/.local/bin/xiaohongshu-*
# 2. 安装系统依赖
apt-get install -y jq xvfb
# 3. 检查依赖
./scripts/install-check.sh
```
### 配置并运行
1. 编辑 `scripts/xhs_trader_monitor.py` 填入你的配置
2. 运行监测:
```bash
cd scripts/
python3 xhs_trader_monitor.py
```
### 配置定时任务
```
5 8,12,16,20 * * * /usr/bin/python3 /path/to/xiaohongshu-trader-monitor/scripts/xhs_trader_monitor.py >> /path/to/logs/xhs_trader.log 2>&1
```
## How It Works
1. **启动 MCP**:自动启动 xiaohongshu-mcp 服务(含 Xvfb 虚拟显示)
2. **获取帖子**:调用 `user_profile` 获取目标用户主页最新帖子
3. **关键词匹配**:检查最近 5 条帖子标题是否含配置的关键词
4. **日期过滤**:只检查**今天**发布的帖子(MM.DD 前缀匹配)
5. **去重推送**:已通知过的帖子 ID 不会重复推送
6. **微信通知**:命中关键词时推送消息到微信
7. **关闭 MCP**:监测完成后自动关闭服务
## Resources
### scripts/
- `xhs_trader_monitor.py` — 核心监测脚本(需配置用户信息)
- `start-mcp.sh` — 启动 MCP 服务(自动管理 Xvfb)
- `stop-mcp.sh` — 停止 MCP 服务
- `mcp-call.sh` — 通用 MCP 调用脚本
- `install-check.sh` — 依赖检查
### references/
- `trader_rules.md` — 完整配置指南、故障排查、消息模板
## Files
| File | Purpose |
|------|---------|
| `scripts/xhs_trader_monitor.py` | Core monitoring script (requires user config) |
| `scripts/start-mcp.sh` | Start MCP service with Xvfb |
| `scripts/stop-mcp.sh` | Stop MCP service |
| `scripts/mcp-call.sh` | Generic MCP tool caller |
| `scripts/install-check.sh` | Dependency checker |
| `references/trader_rules.md` | Configuration guide and troubleshooting |
| `cookies.json` | Login cookies (user-provided, not included) |
| `xhs_trader_state.json` | Notification state (auto-generated) |
don't have the plugin yet? install it then click "run inline in claude" again.