Watermark Removal

This MCP server provides a FastMCP-based service to parse short video links across many platforms, extract the real video URLs without watermarks, and optionally perform video content text extraction. It supports multiple transport methods (stdio, SSE, HTTP) and can be deployed via Docker for scalable use across projects.

How to use

You run the MCP server locally or remotely and connect to it from your MCP client. Use the HTTP endpoint to perform requests via standard HTTP, or use the local stdio interface for fast inter-process communication from scripts or other programs. Two core tools let you parse sharing links and extract optional text content from videos.

How to install

Prerequisites: ensure you have Python installed (recommended Python 3.10+), and FFmpeg installed if you plan to extract video text. You may also use Docker for containerized deployment.

# Install via PyPI (recommended for quick start)
pip install -i https://pypi.org/simple yby6-video-mcp-server

# Verify installation
yby6_video_mcp_server --version

If you prefer to run from source, clone the project, install dependencies, and start the server. Requires Python 3.10+ and a virtual environment.

# From source
git clone https://github.com/yangbuyiya/yby6-crawling-short-video-mcp.git
cd yby6-crawling-short-video-mcp
python -m venv venv
source venv/bin/activate  # on macOS/Linux
venv\Scripts\activate     # on Windows
pip install -r requirements.txt

Docker-based deployment is also supported. Build the base image with FFmpeg and Python, then build the MCP service image and run the container.

# Build base image
docker build -t ffmpeg-python-base:1.0.2 -f Dockerfile.base .

# Build MCP service image
docker build -t yby6-video-mcp:latest -f Dockerfile.mcp .

# Run the MCP service
docker run -d -p 8637:8637 yby6-video-mcp:latest

Additional configuration and endpoints

You can access the server via a quick HTTP/SSE endpoint or use the stdio interface for local development. The following configurations show how to enable either method and how to pass your API key.

{
  "mcpServers": {
    "video_http": {
      "type": "http",
      "name": "video_http",
      "url": "http://127.0.0.1:8637/sse?apikey=xxxxxx",
      "args": []
    }
  }
}

{
  "mcpServers": {
    "video_stdio": {
      "type": "stdio",
      "name": "video_stdio",
      "command": "uv",
      "args": ["yby6_video_mcp_server"],
      "env": {
        "API_KEY": "YOUR_API_KEY"
      }
    },
    "video_stdio_src": {
      "type": "stdio",
      "name": "video_stdio_src",
      "command": "uv",
      "args": [
        "--directory",
        "to/path/yby6-crawling-short-video-mcp/yby6_video_mcp_server",
        "run",
        "-m",
        "yby6_video_mcp_server.server"
      ],
      "env": {
        "API_KEY": "YOUR_API_KEY"
      }
    }
  }
}

How to start using the server

Choose a transport mode and start the server in your chosen environment. If you install via pip, you can run the server with HTTP transport and expose it on port 8637.

# Start via HTTP transport after pip install
yby6_video_mcp_server --transport http --host 0.0.0.0 --port 8637

# Or start from source
python -m yby6_video_mcp_server.server --transport http --host 0.0.0.0 --port 8637

Available tools

share_url_parse_tool

Parses a short video share URL to return the watermark-free video URL, cover image, title, music URL, and author information.

video_id_parse_tool

Parses a video by its source and ID to retrieve the corresponding video data.

share_text_parse_tool

Extracts text content from video by parsing the share link and performing optional transcription using a specified model.