DataHub

Acerca de

Provides an MCP server and a composable Go library designed to integrate AI assistants with DataHub metadata catalogs. It allows AI agents to search datasets, explore schemas, trace data lineage, and access glossary terms and data domains within DataHub. This tool can be deployed as a standalone server for clients like Claude Desktop or Cursor, or used as a Go library to build custom MCP servers with tailored authentication and logging, facilitating powerful metadata-driven interactions for AI.

Características Principales

• Retrieve dataset schemas with field types and descriptions
• Trace upstream and downstream data lineage
• Configure and switch between multiple DataHub server connections
• 1 GitHub stars
• Access glossary term definitions and properties
• Search datasets, dashboards, and pipelines within DataHub

Casos de Uso

• Integrate DataHub metadata capabilities with AI assistants like Claude Desktop or Cursor.
• Develop custom MCP servers with tailored authentication, tenant isolation, and audit logging for DataHub interactions.
• Combine DataHub metadata with other data platforms (e.g., Trino) to build unified data platform MCP servers.