Xiaohongshu MCP

Overview

Xiaohongshu MCP (repository: xpzouying/xiaohongshu-mcp) is a popular open-source Model Context Protocol (MCP) server designed specifically for Xiaohongshu (小红书, also known as RedNote or XHS). It equips AI assistants with powerful tools to automate interactions on the platform using reliable browser automation (Playwright).

The server supports persistent login sessions, content creation and publishing, data retrieval, and various operational tasks — making it one of the most comprehensive MCP servers for Chinese social e-commerce and lifestyle content platforms.

Key Features

• Secure Login Management: One-time QR code login with persistent session saving (avoids frequent re-login).
• Content Publishing: Publish image + text notes (图文笔记), support for videos in some configurations.
• Search & Discovery: Search notes by keywords, fetch personalized recommendation feeds (home feed).
• Interaction Tools: View notes, post comments, monitor notifications.
• Data Retrieval: Extract note details, user information, and more.
• Browser Automation: Runs locally in Chrome/Edge for stability and to avoid IP risks.
• Multi-Platform Support: Pre-built binaries for Windows, macOS, and Linux; also available as Docker image.
• MCP Native: Full compatibility with Claude Desktop, Cursor, Gemini CLI, Cherry Studio, OpenClaw, and other MCP clients.

How It Works

1. Download the binary or Docker image from the repository releases.
2. Run the login script once to authenticate via QR code and save the session.
3. Start the MCP server (default port usually 18060 or configurable).
4. Add the server URL (e.g., http://localhost:18060/mcp) to your MCP client configuration.
5. AI agents can now call tools using natural language, such as "publish a new note with these images and caption" or "search for trending beauty notes".

The server translates MCP tool calls into browser actions on Xiaohongshu.com.

Use Cases

• AI-Powered Content Creation: Generate and auto-publish notes, captions, and visuals on Xiaohongshu.
• Automated Social Operations: Daily posting, commenting, and engagement without manual work.
• Content Research: Search and analyze trending topics, competitor notes, or user feedback.
• Creator Workflows: Schedule or batch-publish lifestyle, fashion, beauty, or e-commerce content.
• Multi-Agent Systems: Combine with other MCP servers for full content pipelines (research → write → publish).

Getting Started

Quick Start (Binary)

• Download the latest release for your OS from GitHub Releases.
• Run the login command to save session.
• Start the MCP server.

Docker

docker run -p 18060:18060 xpzouying/xiaohongshu-mcp

Configure in your MCP client (example for Claude/Cursor):

{
  "mcpServers": {
    "xiaohongshu": {
      "url": "http://localhost:18060/mcp"
    }
  }
}

Detailed guides and troubleshooting are available in the repository README and the author's blog (haha.ai/xiaohongshu-mcp).

Benefits

Xiaohongshu MCP stands out for its stability, ease of use, and focus on real browser simulation, which helps bypass many anti-bot measures. It transforms AI coding assistants into effective Xiaohongshu content operators, supporting creators and marketers who want to automate workflows while keeping operations local and secure.

Note: Always respect Xiaohongshu's terms of service and platform rules when automating. Use responsibly for personal or authorized accounts.

This is currently one of the most actively maintained and widely used MCP servers for Xiaohongshu automation.