playbooks

home / skills / openclaw / skills / xiaohongshu-deep-research

xiaohongshu-deep-research skill

/skills/palmpalm7/xiaohongshu-deep-research

This skill conducts deep Xiaohongshu topic research by crawling top posts, analyzing engagement, and delivering a concise, link-rich summary report.

npx playbooks add skill openclaw/skills --skill xiaohongshu-deep-research

Review the files below or copy the command above to add this skill to your agents.

Files (3)
SKILL.md
2.0 KB
---
name: xiaohongshu-deep-research
description: Deep research on Xiaohongshu (小红书) topics. Use when user wants to research a topic, analyze trends, gather insights from top posts, or produce a summary report. Crawls posts via search, extracts high-engagement content, generates analysis with post links. Requires xiaohongshu-mcp service running.
---

# xiaohongshu-deep-research

小红书话题研究，自动爬取 + 分析 + 输出报告。

## 前置条件

- xiaohongshu-mcp 服务运行中 (`http://localhost:18060/mcp`)
- 已登录小红书账号

## 存储位置

```
~/xiaohongshu-research/{keyword}_{YYYYMMDD_HHmm}/
├── raw/posts.json        # 原始数据
├── analysis/summary.md   # 完整报告
└── metadata.json         # 研究配置
```

## 研究流程

1. **关键词扩展** — 主词 + 3-4 个相关词
2. **批量搜索** — `POST /api/v1/feeds/search`
3. **去重排序** — 按点赞数取 Top 20-50
4. **生成报告** — 含链接、可转发摘要

## 帖子链接格式

```
https://www.xiaohongshu.com/explore/{note_id}
```

## 报告结构

```markdown
# {主题} 小红书研究

---

## 📱 速览（可直接转发）

{2-3段自然语言总结，像朋友分享一样写}

值得看的几篇：
• {标题} — {链接}
• {标题} — {链接}

---

## 数据概览

采集 {N} 篇帖子，共 {likes} 赞 / {collects} 收藏

## Top 10

1. **{标题}** @{作者}
   {likes}👍 {collects}📁 | [查看]({link})

## 发现

{自然语言描述趋势和洞察}

## 数据

原始数据: `{path}`
```

## 写作风格

- 速览部分：像给朋友发消息，口语化，有观点
- 正文部分：简洁，重点突出，避免套话
- 不要用"本报告"、"经过分析"这类官腔

## 限制

⚠️ API 限制：帖子详情和评论无法获取，报告基于搜索结果元数据生成。

## 详细工作流

见 [references/workflow.md](references/workflow.md)

## Credits

基于 [xpzouying/xiaohongshu-mcp](https://github.com/xpzouying/xiaohongshu-mcp)

Overview

This skill performs deep research on Xiaohongshu topics by crawling search results, extracting high-engagement posts, and producing a shareable summary report. It is designed for marketers, creators, and analysts who need quick insights and a curated set of top posts. The service requires the xiaohongshu-mcp endpoint to be running and a logged-in Xiaohongshu account.

How this skill works

The skill expands a seed keyword into related terms, queries the Xiaohongshu search API in bulk, and collects post metadata. It deduplicates and ranks posts by engagement (likes, saves), selects the top items, and writes a concise report with links and a short, friend-style overview. Raw results, metadata, and the final report are stored in a timestamped project folder for review.

When to use it

  • Validate topic demand before launching content campaigns
  • Find top-performing Xiaohongshu posts and creators for partnerships
  • Gather trend signals and audience language for copywriting
  • Produce a quick, shareable research brief for teammates
  • Archive search-based snapshots of a topic over time

Best practices

  • Ensure xiaohongshu-mcp service is running and authenticated before starting
  • Start with a clear seed keyword and allow 3–4 related term expansions
  • Limit results per query to avoid rate limits and focus on Top 20–50 by likes
  • Check the stored raw posts.json when you need original metadata or to re-run analyses
  • Use the friend-style summary for social sharing and the data section for internal decisions

Example use cases

  • Pre-launch content ideation: gather top post formats, tones, and hooks
  • Competitive scan: see what rivals get high engagement on a topic
  • Creator outreach: identify authors of top posts and collect links for pitches
  • Monthly trend report: archive snapshots to compare how interest evolves
  • Copy and creative briefs: extract common phrases and CTAs used in high-engagement notes

FAQ

What does the skill need to run?

The xiaohongshu-mcp service must be available at the configured local endpoint and a Xiaohongshu account must be logged in through that service.

Can it fetch full post text and comments?

No. Due to API limits, the tool uses search result metadata; full post details and comments are not guaranteed.

How are top posts chosen?

Posts are deduplicated and ranked by engagement metrics (likes and collects). The typical output selects the Top 20–50 for the report.