This project is an MCP server that enables AI assistants to search, post, and interact with content on Xiaohongshu (Little Red Book). It provides a set of tools for automating platform operations through MCP-compatible clients, requiring a Xiaohongshu account and network access.
Project overview
It exposes Xiaohongshu operations such as content search, posting, and engagement interactions to AI clients via the Model Context Protocol.
Project type
MCP · AI Agent
Use cases
Automation
Deployment
Refer to project documentation
License
Apache-2.0
Best for
Creators or developers seeking to automate content discovery and platform engagement actions via an MCP-compatible AI client.
Key capabilities
Allows users to log in to Xiaohongshu and verify the current login status.
Search for Xiaohongshu content based on provided keywords.
Retrieve the homepage recommendation list from Xiaohongshu.
Supports automatically posting comments to Xiaohongshu posts.
Like or unlike a post with smart detection of the current interaction state.
Favorite or unfavorite a post with smart detection of the current interaction state.
Reply to specific comments under notes.
Retrieve Xiaohongshu user profile information including basic info and note content.
Limitations and risks
Posting video content only supports local video files and does not support HTTP links.
A Xiaohongshu account does not allow multiple web logins; using the MCP server will disconnect other active web sessions.
There is a risk of account ban if the tools are used for illegal behavior or traffic diversion; the project prohibits illegal actions.
New or unauthenticated accounts may trigger real-name authentication during operation.
Getting started
Download a pre-compiled binary or pull the Docker image, run the login tool to authenticate the Xiaohongshu account, and start the MCP service to connect via an MCP client.
Evidence and sources
README
GitHub project description: MCP for xiaohongshu.com