
Mcp Datahub
Connects Claude to DataHub's GraphQL API for metadata discovery and lineage exploration. Exposes 12 tools covering search, schema inspection, glossary terms, domains, and data lineage tracing. Ships as both a standalone MCP server and a composable Go library, so you can embed it in custom servers with your own auth and tenant logic. Built to work alongside mcp-trino and mcp-s3 for a full data platform stack. Supports multi-instance connections if you run staging and production DataHub clusters. The library includes a QueryProvider interface that lets query engines inject execution context back into metadata responses, turning URNs into queryable table identifiers with sample SQL.
- GitHub stars
- 3
- tools
- 12
- transport
- STDIO
- last updated
- Sep 2, 2026
Tools
12 toolsPublic tool metadata for what this MCP can expose to an agent.
datahub_searchSearch for datasets, dashboards, pipelines by query and entity type
Search for datasets, dashboards, pipelines by query and entity type
No parameter schema in public metadata yet.
datahub_get_entityGet entity metadata by URN (description, owners, tags, domain)
Get entity metadata by URN (description, owners, tags, domain)
No parameter schema in public metadata yet.
datahub_get_schemaGet dataset schema with field types and descriptions
Get dataset schema with field types and descriptions
No parameter schema in public metadata yet.
datahub_get_lineageGet upstream/downstream lineage (supports `level=column` for column-level)
Get upstream/downstream lineage (supports `level=column` for column-level)
No parameter schema in public metadata yet.
datahub_get_queriesGet SQL queries associated with a dataset
Get SQL queries associated with a dataset
No parameter schema in public metadata yet.
datahub_browseBrowse catalog: list tags, domains, or data products
Browse catalog: list tags, domains, or data products
No parameter schema in public metadata yet.
datahub_get_glossary_termGet glossary term definition and properties
Get glossary term definition and properties
No parameter schema in public metadata yet.
datahub_get_data_productGet data product details (owners, domain, properties)
Get data product details (owners, domain, properties)
No parameter schema in public metadata yet.
datahub_list_connectionsList configured DataHub server connections (multi-server mode)
List configured DataHub server connections (multi-server mode)
No parameter schema in public metadata yet.
datahub_createCreate tags, domains, glossary terms, data products, documents, applications, queries, incidents, structured properties, data contracts
Create tags, domains, glossary terms, data products, documents, applications, queries, incidents, structured properties, data contracts
No parameter schema in public metadata yet.
datahub_updateUpdate descriptions (including tag/glossaryTerm descriptions), tags, glossary terms, links, owners, domains, structured properties, custom properties, incidents, queries, documents, data contracts
Update descriptions (including tag/glossaryTerm descriptions), tags, glossary terms, links, owners, domains, structured properties, custom properties, incidents, queries, documents, data contracts
No parameter schema in public metadata yet.
datahub_deleteDelete queries, tags, domains, glossary entities, data products, applications, documents, structured properties
Delete queries, tags, domains, glossary entities, data products, applications, documents, structured properties
No parameter schema in public metadata yet.
Sponsors
Readme
An MCP server and composable Go library that connects AI assistants to DataHub metadata catalogs. Search datasets, explore schemas, trace lineage, and access glossary terms and domains.
Details
- Repository
- github.com/txn2/mcp-datahub
- Category
- Data & Analytics
- Package
https://github.com/txn2/mcp-datahub/releases/download/v1.15.1/mcp-datahub_1.15.1_darwin_arm64.mcpb- Transport
- STDIO
- Registry
- MCP registry, active