Xiaohongshu Toolkit is an MCP server that integrates Xiaohongshu (RED) creator operations with AI clients like Claude Desktop, Cherry Studio, and n8n. It connects local browser automation tools directly to conversational models, enabling content creators, social media managers, and digital marketers to handle publishing workflows and performance tracking entirely through dialogue. Users can publish rich media notes, schedule data collection, and inspect audience analytics without manually logging into the creator web portal. The system uses local Chrome or remote Selenium instances to bypass manual web workflows, executing actions such as posting images and videos, adding topic tags, and pulling CSV-based dashboard statistics. With structured Chinese headers natively formatted for AI comprehension, connected language models can inspect impressions, likes, and follower growth trends directly. While the original repository is marked as no longer maintained by its author, the toolkit provides complete automation mechanisms for self-hosted instances seeking custom conversational management over Xiaohongshu creator channels.
Category: CRM, ERP & E-commerce
Tags: content-creation, marketing, social-media, xiaohongshu
bash git clone https://github.com/aki66938/xhs-toolkit.git cd xhs-toolkit cp env_example .env 2. Open .env and configure your Chrome and ChromeDriver paths: bash CHROME_PATH="/path/to/Google Chrome" WEBDRIVER_CHROME_DRIVER="/path/to/chromedriver" 3. Install dependencies using uv: bash uv sync 4. Save your login cookies via the interactive menu: bash ./xhs cookie save 5. Add the server configuration to your Claude Desktop config file (claude_desktop_config.json): json { "mcpServers": { "xhs-toolkit": { "command": "uv", "args": [ "--directory", "/path/to/xhs-toolkit", "run", "python", "-m", "src.server.mcp_server", "--stdio" ] } } } Replace /path/to/xhs-toolkit with your absolute local directory path, then restart Claude Desktop.Part of MCP Servers
Xiaohongshu Toolkit connects conversational AI tools directly to the Xiaohongshu creator platform. It automates cookie management, publishes text, image, and video notes with topic tags, and extracts creator dashboard and follower analytics into local CSV files for LLM review.
The toolkit works with any client supporting standard Model Context Protocol integrations over stdio, including Claude Desktop, Cherry Studio, and automation platforms like n8n using an AI agent node configured with external MCP tools.
Clone the repository, set your Chrome and ChromeDriver paths in the environment file, and install dependencies using uv sync or pip. Run the initial cookie save command to log into the creator center, then register the stdio command in your client configuration.
According to the repository documentation, the original author has stopped updating the project and does not plan further maintenance. Users can fork the repository to modify signature algorithms, update automation logic, and run custom builds.