hk-health

This MCP server, hk-health, exposes Hong Kong health data through a FastMCP interface. It provides three main data capabilities: current Accident and Emergency Department waiting times by Hong Kong hospital, current wait times for new case bookings for specialist outpatient services by specialty and cluster, and the average number of general outpatient clinic quotas across 18 districts for the preceding four weeks. The server is implemented in Python and can run in standard I/O mode or in Server-Sent Events (SSE) mode. To use it in standard I/O, start the server with python server.py, which will listen on the default I/O channel. For streaming updates via SSE, start the server with python server.py --sse (SSE mode typically serves on port 8000). The README also shows how to connect to the MCP using standard IO in integration scenarios such as Cline.

Prerequisites:

• Python 3.8+ installed on your system
• Git to clone the repository
• Network access to install Python dependencies

Step-by-step installation:

1. Clone the repository: git clone https://github.com/hkopenai/hk-health-mcp-server.git
2. Navigate to the project directory: cd hk-health-mcp-server
3. Install Python dependencies: pip install -r requirements.txt
4. Run the server (default I/O mode): python server.py (Optional) Run in SSE mode: python server.py --sse

If you prefer to use a virtual environment:

• python -m venv venv
• source venv/bin/activate # on Unix
• .\venv\Scripts\activate # on Windows
• pip install -r requirements.txt

Notes and tips:

• Data is sourced from the Hong Kong Hospital Authority.
• SSE mode serves streaming data; use --sse to enable it (default port is 8000 for SSE).
• Cline integration can connect via stdio using an external MCPl-compatible runner, as shown in the README snippet. The provided configuration uses uvx for stdio integration.
• If you encounter import or dependency issues, ensure your Python environment has access to the requirements specified in requirements.txt and that the network access to the data sources is allowed.
• For debugging, run in a virtual environment and enable verbose logging if available in server.py.