This MCP server suite provides a complete AI-powered social media operating platform focused on Xiaohongshu (Little Red Book). It is composed of six MCP services: a content generation engine, an image generator, a video generator, a browser automation component for platform operations, a data collector for analytics, and the AI scheduler that coordinates tasks and strategies. Each service exposes an MCP API and can be run independently via uvx. Start by launching the image and content services to generate assets and drafts, then use the browser automation MCP to publish and manage posts on Xiaohongshu. The AI scheduler ties everything together, driving autonomous decision making, scheduling, and interaction between services to automate your workflow.

Prerequisites:

• Python 3.11 or higher
• uvx package manager installed on your system
• Git to clone repositories

Step-by-step installation:

1. Clone the repository (or ensure you have the project structure): git clone https://github.com/luyike221/xiaohongshu-mcp-python.git cd xiaohongshu-mcp-python

2. Install UV package manager (example commands from README):

   macOS/Linux

   curl -LsSf https://astral.sh/uv/install.sh | sh

   Windows

   powershell -c "irm https://astral.sh/uv/install.ps1 | iex"

3. Install individual MCP services dependencies (as per README):

   Image generation MCP

   cd xhs-image-mcp uv sync

   Video generation MCP

   cd ../xhs-video-mcp uv sync

   Content generation MCP

   cd ../xhs-content-generator-mcp uv sync

   Browser automation MCP

   cd ../xhs-browser-automation-mcp uv sync uv run playwright install chromium

   Data collector MCP

   cd ../xhs-data-collector-mcp uv sync

   AI scheduler

   cd ../ai_social_scheduler uv sync

4. Configure each service by creating .env files as described in the README for environment-specific settings.

5. Start services (examples):

   Image generation

   cd xhs-image-mcp uv run python -m image_video_mcp.main

   Service will be available at http://127.0.0.1:8003

   Video generation

   cd ../xhs-video-mcp ./start.sh # or uv run xhs-video-mcp

   Default: http://localhost:8005

   Content generation

   cd ../xhs-content-generator-mcp uv run python -m xhs_content_generator_mcp.main

   Browser automation

   cd ../xhs-browser-automation-mcp uv run python -m main

   Data collector

   cd ../xhs-data-collector-mcp uv run python -m data_collector_mcp.main

   AI scheduler

   cd ../ai_social_scheduler uv run python -m scheduler.main

Tips and common issues:

• Ensure the correct Python version (3.11+) is used and that uvx is installed before running uv commands.
• Some MCPs require API keys (e.g., image generation keys, OpenAI) to be set in .env files; never commit these keys to version control.
• For the video MCP, GPU acceleration requires compatible hardware and drivers; set VIDEO_GPU_ACCELERATION to false if not available.
• When starting multiple MCP services, verify that port mappings do not conflict (adjust MCP_PORT as needed).
• The ai_social_scheduler relies on multiple other MCP services; start dependencies first, then the scheduler.
• If a service fails to start, check the corresponding .env for required variables and ensure network access to external APIs.
• The npm_package field is not applicable for this Python-based MCP suite; use the Python uvx commands to manage services.