MCP Trino Server

What is MCP Trino Server?

MCP Trino Server is a Model Context Protocol (MCP) server that provides seamless integration with Trino and Iceberg, enabling advanced data exploration, querying, and table maintenance through a standard interface. It is designed for developers and data engineers building AI‑powered tools that interact with Trino databases.

How to use MCP Trino Server?

Install via uv, pip, or Docker, then configure using environment variables (TRINO_HOST, TRINO_PORT, TRINO_USER, etc.). Run the server with python src/server.py (or the Docker equivalent) using one of three transport modes: stdio (default), streamable‑http, or sse. Connect an MCP client (e.g., VS Code or Claude Desktop) by providing the server’s command and configuration.

Key features of MCP Trino Server

• Execute SQL queries with formatted results
• Browse catalogs, schemas, and tables hierarchically
• Perform Iceberg table maintenance (optimize, expire snapshots)
• Inspect Iceberg metadata (snapshots, manifests, partitions, files)
• Supports stdio, streamable‑HTTP, and SSE transports
• Configurable via environment variables or .env file

Use cases of MCP Trino Server

• Interactive data exploration and analysis in Trino
• Automated Iceberg table maintenance and optimization
• Building AI‑powered tools that interact with Trino databases
• Executing and managing SQL queries with proper result formatting

FAQ from MCP Trino Server

What are the prerequisites to run MCP Trino Server?

You need Python 3.12 or higher and a running Trino server (Docker Compose is available for local development). Docker is optional but recommended for containerized deployment.

What transport modes are supported?

The server supports stdio (default, for local MCP clients like VS Code and Claude Desktop), streamable‑http (for remote or web access), and sse (Server‑Sent Events, legacy HTTP transport).

How do I configure the server?

Configure via environment variables: TRINO_HOST, TRINO_PORT, TRINO_USER, TRINO_CATALOG, TRINO_SCHEMA, TRINO_HTTP_SCHEME, and optionally TRINO_PASSWORD. You can also use a .env file.

Does the server support authentication to Trino?

Yes, you can set the TRINO_PASSWORD environment variable to provide a password for authentication to the Trino server.

Can I deploy the server in a container?

Yes. Build the Docker image with docker build -t mcp-trino-python . and run it with either stdio or streamable‑http transport. Docker Compose profiles are also provided for quick start.