Iceberg MCP

Name: Iceberg MCP
Availability: InStock
Author: morristai

by morristai

Connect your AI to Apache Iceberg catalogs for high-performance lakehouse data exploration.

0 stars

3 views

Works in:claude

Exposes:Tools

View on GitHub Docs

What it does

Iceberg MCP provides a bridge between Large Language Models and Apache Iceberg catalogs. It allows AI agents to discover, inspect, and explore data stored in an Iceberg lakehouse architecture, supporting both REST catalogs and AWS Glue.

Tools

namespaces: Retrieves all available namespaces in the Iceberg catalog.
tables: Lists all tables within a specific namespace.
table_schema: Returns the detailed schema for a given table.
table_properties: Retrieves the configuration and properties of a specific table.

Installation

To integrate with Claude Desktop, add the following to your claude_desktop_config.json:

{
  "mcpServers": {
    "iceberg-mcp": {
      "command": "PATH-TO-BINARY/iceberg-mcp",
      "env": {
        "CATALOG_KIND": "rest",
        "REST_URI": "http://localhost:8080",
        "LOG_LEVEL": "info"
      }
    }
  }
}

(Note: AWS Glue configuration is also available using CATALOG_KIND: glue and AWS credentials)