iceberg-mcp

iceberg-mcp

37

IcebergMCP is an AI-native lakehouse integration tool that leverages the Model Context Protocol to interact with Apache Iceberg data using natural language through various MCP clients. It provides read-only insights into data structures, like tables and schemas, in AWS Glue or planned Iceberg REST Catalog environments.

IcebergMCP 🚀

AI-native Lakehouse Integration

IcebergMCP is a Model Context Protocol (MCP) server enabling natural language interaction with Apache Iceberg Lakehouse through clients like Claude or Cursor.

Features

  • Interacts with Apache Iceberg tables using natural language.
  • Supports tools to get namespaces, tables, schema, properties, and partitions.

Installation

  • Requires Apache Iceberg catalog on AWS Glue and AWS profile setup.
  • uv package manager needed for operation.

Configuration

  • Uses environment variables for AWS connection, such as ICEBERG_MCP_PROFILE and ICEBERG_MCP_REGION.

Examples

  • Interactions include fetching namespaces, table details, schema, and more.

Limitations & Security

  • Tools are read-only and cannot modify data.
  • Currently supports AWS Glue and soon Apache Iceberg REST Catalog.

Contributing

Contributions via pull requests are welcome.