A Model Context Protocol (MCP) service for programmatically managing Xiaohongshu accounts, content publishing, and user engagement.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install xiaohongshu-mcp-skill
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install xiaohongshu-mcp-skill using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Xiaohongshu MCP?
The Xiaohongshu MCP is a specialized integration designed to bridge the gap between AI agents and the popular Chinese social media platform, Xiaohongshu (RED). By leveraging this service within the broader ecosystem of Openclaw Skills, developers and marketers can programmatically handle tasks that typically require manual mobile or browser interaction. The skill operates as a local Model Context Protocol service, allowing for low-latency communication between your AI agent and the Xiaohongshu platform.
This tool is particularly valuable for users who need to scale their presence on RED. Whether you are managing multiple accounts or conducting large-scale market research, incorporating this into your Openclaw Skills library provides the technical infrastructure to automate content discovery, interaction, and publishing with precision.
Xiaohongshu MCP Use Cases
- Automating the publication of image-based and video-based notes to Xiaohongshu.
- Conducting automated content research by searching feeds for specific keywords or trends.
- Managing community engagement through automated post likes, favorites, and threaded comments.
- Monitoring account health and login status across multiple profiles programmatically.
- Building custom dashboards that aggregate Xiaohongshu feed details and user profile data.
How Xiaohongshu MCP Works
- The user deploys the local MCP binary and ensures the service is listening on port 18060.
- The AI agent initializes the connection via a JSON-RPC call to the /mcp endpoint.
- Authentication is handled by retrieving a login QR code or verifying existing session cookies via the check_login_status tool.
- Once authenticated, the agent can call specialized functions like search_feeds or publish_content to interact with the platform.
- The MCP service translates these agent requests into platform-specific commands and returns structured data to the agent for further processing.
Xiaohongshu MCP Setup
To begin using this skill with Openclaw Skills, follow these installation and verification steps:
- Verify if the MCP service is already running on your local machine:
curl -s -X POST http://localhost:18060/mcp \
-H "Content-Type: application/json" \
-d '{"jsonrpc":"2.0","method":"initialize","params":{},"id":1}'
- If the service is not running, navigate to the binary directory and start it in the background:
cd /your/local/path/xiaohongshu-mcp
nohup ./bin/xiaohongshu-mcp-darwin-arm64 > mcp.log 2>&1 &
- Ensure your AI agent is configured to point to the localhost:18060 address to access the toolset.
Xiaohongshu MCP Data Schema & Taxonomy
The skill organizes its operations through a set of structured tools and data models. Understanding these limits is crucial for maintaining account standing when using Openclaw Skills:
| Component | Constraint / Details |
|---|---|
| Title Length | Maximum 20 characters |
| Content Length | Maximum 1000 characters |
| Daily Publishing Limit | 50 posts per day |
| Session Data | Managed via local cookies; can be reset using delete_cookies |
| Feed Data | Returns structured note details including IDs, descriptions, and user info |
Xiaohongshu MCP Advanced Features
- Multi-format publishing support for both static image notes and dynamic video content.
- Comprehensive engagement suite including the ability to reply to specific comments within a feed.
- Automated session recovery and cookie management to ensure persistent agent access.
- Content discovery tools that allow for deep searching and retrieval of specific note details for sentiment analysis.
- Optimized for local execution, ensuring sensitive login data remains within your controlled environment when using Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
Qwen TTS (Local AI Voice Synthesis)
A powerful local text-to-speech engine powered by Qwen3-TTS for generating emotional and stylistically controlled audio.
ClawBoss
ClawBoss is an AI-powered productivity coach that utilizes professional coaching frameworks like the GROW model to guide goal setting and maintain momentum.
GeeLark Complete
An end-to-end automation skill for GeeLark providing API orchestration, native desktop synchronization, and RPA fallback capabilities.
Molta AI Q&A Platform
Molta is a specialized Q&A platform designed for AI agents to register, verify, and participate in collaborative knowledge sharing.
Azure DevOps
A powerful integration for managing Azure DevOps projects, repositories, and pull request lifecycles directly through an AI-driven interface.
ARTA — Agentic Real-Time Awareness
ARTA provides an in-memory awareness layer for AI agents to track and query their own activities across multiple sessions within a single process.
Featured*
