RedNote Analyzer MCP - 小红书 AI 分析工具

Install: pipx install rednote-analyzer-mcp | PyPI: pypi.org/project/rednote-analyzer-mcp

MCP Server for Xiaohongshu (小红书/RED) - 让 Claude、ChatGPT 等 AI 助手能够搜索、分析、生成小红书内容。

A Model Context Protocol server that enables AI assistants to search, analyze, and generate content for China's leading lifestyle social media platform.

Keywords: 小红书 API, 小红书爬虫, 小红书数据分析, XHS API, RedNote API, MCP Server, Claude Tools, AI Content Generation, 小红书笔记分析, 爆款分析

English | 中文

---

先决条件

• Python 3.11+
• pipx 或 uv

安装

第一步：安装

pipx install rednote-analyzer-mcp

第二步：登录小红书（必须）

rednote-login

浏览器会打开小红书，扫码或手机号登录，登录成功后自动保存 Cookie。

⚠️ 不登录直接用会返回空结果！

⚠️ 风控提示：小红书有严格的反爬机制。本工具已内置限流（每分钟最多 10 次请求，每次请求间隔至少 3 秒），但仍建议：

• 避免短时间内大量请求
• 如果账号被限制，等待 24 小时后再试
• 仅用于个人学习研究，不要用于商业爬取

第三步：配置 AI 工具

选择你用的工具：

Claude Code

在项目根目录创建 .mcp.json 文件：

{
  "mcpServers": {
    "rednote-analyzer-mcp": {
      "command": "uvx",
      "args": ["rednote-analyzer-mcp"],
      "env": {
        "REDNOTE_ADAPTER": "playwright",
        "REDNOTE_HEADLESS": "true"
      }
    }
  }
}

或者用命令添加：

# 添加到当前项目
claude mcp add rednote-analyzer-mcp -s local \
  -e REDNOTE_ADAPTER=playwright \
  -e REDNOTE_HEADLESS=true \
  -- uvx rednote-analyzer-mcp

# 添加到全局（所有项目可用）
claude mcp add rednote-analyzer-mcp -s user \
  -e REDNOTE_ADAPTER=playwright \
  -e REDNOTE_HEADLESS=true \
  -- uvx rednote-analyzer-mcp

Claude Desktop

编辑 claude_desktop_config.json：

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "rednote-analyzer-mcp": {
      "command": "uvx",
      "args": ["rednote-analyzer-mcp"],
      "env": {
        "REDNOTE_ADAPTER": "playwright",
        "REDNOTE_HEADLESS": "true"
      }
    }
  }
}

Cursor

创建 .cursor/mcp.json（项目级）或 ~/.cursor/mcp.json（全局）：

{
  "mcpServers": {
    "rednote-analyzer-mcp": {
      "command": "uvx",
      "args": ["rednote-analyzer-mcp"],
      "env": {
        "REDNOTE_ADAPTER": "playwright",
        "REDNOTE_HEADLESS": "true"
      }
    }
  }
}

VS Code (GitHub Copilot)

创建 .vscode/mcp.json：

{
  "servers": {
    "rednote-analyzer-mcp": {
      "command": "uvx",
      "args": ["rednote-analyzer-mcp"],
      "env": {
        "REDNOTE_ADAPTER": "playwright",
        "REDNOTE_HEADLESS": "true"
      }
    }
  }
}

Windsurf

编辑 ~/.codeium/mcp_config.json：

{
  "mcpServers": {
    "rednote-analyzer-mcp": {
      "command": "uvx",
      "args": ["rednote-analyzer-mcp"],
      "env": {
        "REDNOTE_ADAPTER": "playwright",
        "REDNOTE_HEADLESS": "true"
      }
    }
  }
}

其他工具

通用配置模式：

| 配置项 | 值 | |-------|---| | command | uvx | | args | ["rednote-analyzer-mcp"] | | env | {"REDNOTE_ADAPTER": "playwright", "REDNOTE_HEADLESS": "true"} |

---

使用

配置完成后，直接跟 AI 对话：

"搜索小红书上关于美股的热门笔记"

"分析这些笔记的爆款规律"

"帮我写一篇关于基金定投的小红书文案"

"把这段文字改写成小红书风格"

---

工具列表

| 工具 | 功能 | |------|------| | rednote_search_notes | 搜索笔记 | | rednote_get_note_detail | 获取笔记详情和评论 | | rednote_analyze_note | 分析笔记结构和爆款元素 | | rednote_extract_patterns | 批量分析，提取规律 | | rednote_generate_post | 生成帖子大纲 | | rednote_rewrite_in_style | 改写为小红书风格 |

---

常见问题

Q: 怎么更新？

pipx upgrade rednote-analyzer-mcp

Q: 搜索返回空结果？

A: 没有登录。运行 rednote-login 登录一次。

Q: Cookie 过期了？

A: 重新运行 rednote-login。

Q: 怎么查看 MCP 是否连接成功？

A: Claude Code 运行 claude mcp list，其他工具看各自的 MCP 状态面板。

---

License

MIT | 开发指南

---

English

What is RedNote Analyzer MCP?

RedNote Analyzer MCP is a Model Context Protocol (MCP) server that connects AI assistants like Claude to Xiaohongshu (小红书/RED/RedNote), China's most popular lifestyle social media platform with 300M+ users.

Use Cases

• Content Research: Search and analyze trending posts on any topic
• Viral Pattern Analysis: Extract what makes content go viral on RedNote
• Content Generation: Generate RedNote-style posts with AI assistance
• Market Research: Understand Chinese consumer trends and preferences

Prerequisites

• Python 3.11+
• pipx or uv
• Chromium browser (auto-downloaded during install)

Installation

Step 1: Install

pipx install rednote-analyzer-mcp

Step 2: Log in to Xiaohongshu (required)

rednote-login

A browser will open. Log in with your phone number or scan QR code. Cookies are saved automatically after login.

⚠️ Skipping login will result in empty search results!

⚠️ Risk Warning: Xiaohongshu has strict anti-scraping measures. This tool has built-in rate limiting (max 10 requests/minute, min 3 seconds between requests), but please:

• Avoid making too many requests in a short time
• If your account gets restricted, wait 24 hours before retrying
• Use only for personal learning and research, not commercial scraping

Step 3: Configure your AI tool

Choose your tool:

Claude Code

Create .mcp.json in your project root:

{
  "mcpServers": {
    "rednote-analyzer-mcp": {
      "command": "uvx",
      "args": ["rednote-analyzer-mcp"],
      "env": {
        "REDNOTE_ADAPTER": "playwright",
        "REDNOTE_HEADLESS": "true"
      }
    }
  }
}

Or use CLI:

# Add to current project
claude mcp add rednote-analyzer-mcp -s local \
  -e REDNOTE_ADAPTER=playwright \
  -e REDNOTE_HEADLESS=true \
  -- uvx rednote-analyzer-mcp

# Add globally (all projects)
claude mcp add rednote-analyzer-mcp -s user \
  -e REDNOTE_ADAPTER=playwright \
  -e REDNOTE_HEADLESS=true \
  -- uvx rednote-analyzer-mcp

Claude Desktop

Edit claude_desktop_config.json:

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "rednote-analyzer-mcp": {
      "command": "uvx",
      "args": ["rednote-analyzer-mcp"],
      "env": {
        "REDNOTE_ADAPTER": "playwright",
        "REDNOTE_HEADLESS": "true"
      }
    }
  }
}

Cursor

Create .cursor/mcp.json (project) or ~/.cursor/mcp.json (global):

{
  "mcpServers": {
    "rednote-analyzer-mcp": {
      "command": "uvx",
      "args": ["rednote-analyzer-mcp"],
      "env": {
        "REDNOTE_ADAPTER": "playwright",
        "REDNOTE_HEADLESS": "true"
      }
    }
  }
}

VS Code (GitHub Copilot)

Create .vscode/mcp.json:

{
  "servers": {
    "rednote-analyzer-mcp": {
      "command": "uvx",
      "args": ["rednote-analyzer-mcp"],
      "env": {
        "REDNOTE_ADAPTER": "playwright",
        "REDNOTE_HEADLESS": "true"
      }
    }
  }
}

Windsurf

Edit ~/.codeium/mcp_config.json:

{
  "mcpServers": {
    "rednote-analyzer-mcp": {
      "command": "uvx",
      "args": ["rednote-analyzer-mcp"],
      "env": {
        "REDNOTE_ADAPTER": "playwright",
        "REDNOTE_HEADLESS": "true"
      }
    }
  }
}

Other Tools

General config pattern:

| Key | Value | |-----|-------| | command | uvx | | args | ["rednote-analyzer-mcp"] | | env | {"REDNOTE_ADAPTER": "playwright", "REDNOTE_HEADLESS": "true"} |

---

Usage

After setup, talk to your AI:

"Search for trending posts about US stocks (美股) on RedNote"

"Analyze the viral patterns of these notes"

"Help me write a RedNote post about ETF investing"

"Rewrite this text in RedNote style"

---

Tools

| Tool | Function | |------|----------| | rednote_search_notes | Search notes | | rednote_get_note_detail | Get note details and comments | | rednote_analyze_note | Analyze note structure and viral elements | | rednote_extract_patterns | Batch analyze, extract patterns | | rednote_generate_post | Generate post outline | | rednote_rewrite_in_style | Rewrite in RedNote style |

---

FAQ

Q: How to update?

pipx upgrade rednote-analyzer-mcp

Q: Search returns empty results?

A: You haven't logged in. Run rednote-login to log in.

Q: Cookie expired?

A: Run rednote-login again.

Q: How to check if MCP is connected?

A: Claude Code: run claude mcp list. Other tools: check their MCP status panel.

---

License

MIT | Contributing