dolphindb

This MCP server integrates DolphinDB as a data backend and exposes FastMCP-compatible functions for exploring and querying DolphinDB databases and tables. After installation, start the server to launch the MCP plugin service. Once running, you can access the available capabilities through the MCP frontend or your language toolchain. The server exposes functions such as list_dbs(), list_tbs(dbName: str), query_table_diskusage(database: str, tableName: str), and query_dolphindb(script: str). These allow you to discover DolphinDB databases, inspect tables, check storage usage, and run DolphinDB scripts via the MCP interface.

Prerequisites:

• Python 3.8+
• pip
• Optional: virtual environment tool (venv)

Installation methods:

Method 1 — uvx (recommended, if you're distributing via UV Universe):

uvx install dolphindb-mcp-server

Run:

uvx dolphindb-mcp-server

Note: uvx installation requires packaging with uv.pack and uploading to Universe. See UV docs for packaging and publishing.

Method 2 — PyPI install

pip install dolphindb-mcp-server

Run:

dolphindb-mcp-server

Method 3 — Local build & install

git clone https://github.com/your-org/dolphindb-mcp-server.git
cd dolphindb-mcp-server

# Create and activate virtual environment
python -m venv .venv
source .venv/bin/activate

# Build and install
pip install build
python -m build
pip install dist/*.whl

Run:

dolphindb-mcp-server

Environment variables control DolphinDB connection during startup. If not set, defaults will be used (host=127.0.0.1, port=8848, user=admin, password=123456). Ensure DolphinDB server is reachable from the MCP server host. If you deploy in production, consider securing credentials and using a vault or environment-specific config. If the MCP server startup fails, check DolphinDB connectivity, Python version, and that the entrypoint module dolphindb_mcp_server is available in your PYTHONPATH.