MCP Server Trino

🚀 Trino MCP Server

This repository provides an MCP (Model Control Protocol) server that allows you to list and query tables using Trino via Python.

🚀 Quick Start

This README provides a comprehensive guide to the Trino MCP Server, including its overview, requirements, configuration, and usage examples.

✨ Features

• MCP: MCP is a protocol for bridging AI models, data, and tools. This example MCP server offers the following capabilities:
  • List Trino tables as MCP resources.
  • Read table contents via MCP.
  • Tools to execute arbitrary SQL queries to access Trino data.
• Trino: A fast, distributed SQL query engine for big data analytics. This server leverages the Trino Python client (trino.dbapi) to connect to Trino hosts, catalogs, and schemas.

📦 Installation

Requirements

• Python 3.9+ (or a version compatible with mcp, trino, and asyncio).
• trino (Trino's Python driver).
• mcp (Python Model Control Protocol library).

📚 Documentation

Configuration

The server reads Trino connection details from environment variables:

💻 Usage Examples

Basic Usage

{
  "mcpServers": {
    "trino": {
      "command": "uv",
      "args": [
        "--directory", 
        "<path_to_mcp_server_trino>",
        "run",
        "mcp_server_trino"
      ],
      "env": {
        "TRINO_HOST": "<host>",
        "TRINO_PORT": "<port>",
        "TRINO_USER": "<user>",
        "TRINO_PASSWORD": "<password>",
        "TRINO_CATALOG": "<catalog>",
        "TRINO_SCHEMA": "<schema>"
      }
    }
  }
}

This JSON configuration example demonstrates how to configure the Trino service in the MCP server, including setting command-line arguments and environment variables. Each field needs to be replaced with actual values to ensure a proper connection to the target Trino server.