explaingit

xpzouying/xiaohongshu-mcp

13,570GoAudience · developerComplexity · 3/5Setup · moderate

TLDR

A connector that lets AI assistants automate Xiaohongshu (Little Red Book), publishing posts, searching content, reading feeds, commenting, liking, and bookmarking, using the MCP standard interface.

Mindmap

mindmap
  root((xiaohongshu-mcp))
    What it does
      Publishes posts
      Searches content
      Reads comments
      Likes and bookmarks
    Deployment options
      Browser extension
      Pre-built binary
      Docker container
    Platforms
      macOS
      Windows
      Linux
    Protocol
      MCP standard
      AI assistant bridge
Click or tap to explore — scroll the page freely

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

Things people build with this

USE CASE 1

Connect an AI assistant to a Xiaohongshu account to automatically publish posts with images or video through conversation

USE CASE 2

Use an AI assistant to search Xiaohongshu for trending keywords and retrieve post details including engagement numbers

USE CASE 3

Automate replying to comments on Xiaohongshu posts through an AI assistant using the MCP interface

Tech stack

GoDockerMCP

Getting it running

Difficulty · moderate Time to first run · 30min

Requires an active Xiaohongshu account and an MCP-compatible AI client such as Claude Desktop, the same account cannot be signed into multiple sessions simultaneously.

In plain English

This repository provides a connector that lets AI assistants interact directly with Xiaohongshu, also known as Little Red Book or RED, a Chinese social media platform focused on lifestyle content, short videos, and product reviews. The connector follows the MCP protocol, a standard interface for giving AI tools access to external services. The project is written in Go. The primary purpose is automation: once connected, an AI assistant can log into a Xiaohongshu account, publish posts containing images or video, search for content by keyword, retrieve the recommendation feed, read post details including comments and engagement numbers, post comments, reply to specific users, like or unlike posts, and bookmark content. The README states that all of these features were tested running stably for several weeks before the code was released. Three deployment options are provided. The simplest is a browser extension called x-mcp that requires no server setup at all. The second is a pre-compiled binary downloaded from the releases page, with builds available for macOS (Intel and Apple Silicon), Windows, and Linux. The third option is a Docker container that can be pulled and run with a single command. Building from source using Go is also supported. The README, written primarily in Chinese, includes practical notes on Xiaohongshu's own content rules: post titles must be under 20 characters, body text must be under 1,000 characters, and the same account cannot be signed into multiple browser sessions simultaneously without being logged out of the MCP session. The project notes that all donations received from supporters are given to charitable organizations. A caution is included stating that the project is intended for learning purposes and that illegal uses are prohibited. The full README is longer than what was shown.

Copy-paste prompts

Prompt 1
Set up xiaohongshu-mcp using the pre-compiled binary for macOS, connect it to Claude Desktop, then write a prompt that searches for the keyword 'skincare routine', retrieves the top 5 posts, and summarizes the key tips mentioned
Prompt 2
Using xiaohongshu-mcp connected to my AI assistant, write a prompt that publishes a new Xiaohongshu post with a title under 20 characters and body text under 1000 characters promoting a product, and confirm the post was published successfully
Prompt 3
Pull the xiaohongshu-mcp Docker container, connect it to an MCP client, and write a prompt that fetches my recommendation feed and identifies the 3 posts with the highest engagement numbers
Prompt 4
Using xiaohongshu-mcp, write a prompt that retrieves all comments on my 5 most recent posts and automatically replies to any comment that contains a question mark
Open on GitHub → Explain another repo

← xpzouying on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.