Iceberg

Provides read-only access to Iceberg tables in Impala, enabling schema inspection and read-only queries.

python

GitHub Stars

python

Language

6 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "peterableda-iceberg-mcp-server-original": {
      "command": "uvx",
      "args": [
        "--from",
        "git+https://github.com/cloudera/iceberg-mcp-server@main",
        "run-server"
      ],
      "env": {
        "IMPALA_HOST": "coordinator-default-impala.example.com",
        "IMPALA_PORT": "443",
        "IMPALA_USER": "username",
        "IMPALA_DATABASE": "default",
        "IMPALA_PASSWORD": "password"
      }
    }
  }
}

This MCP server provides read-only access to Iceberg tables through Apache Impala, enabling you to inspect database schemas and run read-only queries. It exposes simple endpoints you can call from your MCP client or AI framework to retrieve schemas and execute queries against Impala.

How to use

You can connect to the Iceberg MCP Server from your MCP client or AI framework to perform two main actions: list available tables in the current database and run read-only SQL queries against Impala. Choose one of the provided deployment options, configure the required environment variables, and start the server. Once running, you can invoke the server’s read-only endpoints to inspect schemas and fetch query results as JSON.

Using with Claude Desktop or MCP clients

Option 1 Direct installation from GitHub (Recommended) provides a self-contained runtime that starts the server from a git source and runs in a local process. Use the following configuration snippet in your MCP client configuration to start the server.

{
  "mcpServers": {
    "iceberg-mcp-server": {
      "command": "uvx",
      "args": [
        "--from",
        "git+https://github.com/cloudera/iceberg-mcp-server@main",
        "run-server"
      ],
      "env": {
        "IMPALA_HOST": "coordinator-default-impala.example.com",
        "IMPALA_PORT": "443",
        "IMPALA_USER": "username",
        "IMPALA_PASSWORD": "password",
        "IMPALA_DATABASE": "default"
      }
    }
  }
}

"} ,{

## Option 2 Local installation after cloning the repository

Option 2 runs the MCP server from a local directory after you clone the repository. Use this when you prefer to run the server from your own environment.

{ "mcpServers": { "iceberg-mcp-server": { "command": "uv", "args": [ "--directory", "/path/to/iceberg-mcp-server", "run", "src/iceberg_mcp_server/server.py" ], "env": { "IMPALA_HOST": "coordinator-default-impala.example.com", "IMPALA_PORT": "443", "IMPALA_USER": "username", "IMPALA_PASSWORD": "password", "IMPALA_DATABASE": "default" } } } }

"} ,{

Environment and transport options

The server runs with Impala connection details provided via environment variables. You can adapt these values to your Impala setup.

Usage with AI frameworks

The server supports integration with common AI frameworks. It exposes endpoints that return query results as JSON and a schema listing that helps you discover available tables.

Transport options let you choose how the server communicates with your tooling. The following transport modes are supported.

Transport options

The MCP server’s transport protocol is configurable via the MCP_TRANSPORT environment variable. Supported values are stdio, http, and sse.

Available tools

execute_query

Run a SQL query against Impala and return the results as JSON.

get_schema

List all tables available in the current database.

Built by

VeilStrat

AI signals for GTM teams