DataHub MCP Server
Official DataHub MCP server enabling AI agents to interact with DataHub — the open-source metadata management and data catalog platform — searching data assets, querying lineage graphs, accessing schema information, and integrating DataHub's comprehensive metadata catalog into agent-driven data discovery and governance workflows.
Score Breakdown
⚙ Agent Friendliness
🔒 Security
HTTPS. Access token. Official MCP from acryldata. Metadata catalog may be sensitive.
⚡ Reliability
Best When
A data team uses DataHub as their metadata catalog and wants AI agents to understand data assets — enabling agents to discover, understand, and reason about available data.
Avoid When
You use Amundsen, Alation, or another data catalog — use the appropriate MCP. DataHub is for metadata discovery, not data access.
Use Cases
- • Searching the data catalog for relevant datasets from data discovery agents
- • Querying data lineage for pipeline debugging from data engineering agents
- • Retrieving schema information for data assets from analysis agents
- • Checking data quality and freshness from monitoring agents
- • Finding dataset owners and stewards from governance agents
- • Building data-aware RAG systems with metadata context from AI agents
Not For
- • Teams not using DataHub for data cataloging
- • Direct data querying (DataHub provides metadata, not data access)
- • Teams using Amundsen, Alation, or other data catalogs
Interface
Authentication
DataHub personal access token required. Generate from DataHub UI Settings → Access Tokens. DataHub GMS URL must be configured.
Pricing
DataHub is free open source. Official MCP from acryldata (DataHub creator). Cloud version for managed service.
Agent Metadata
Known Gotchas
- ⚠ DataHub deployment (GMS server) must be running and accessible
- ⚠ DataHub's metadata model is rich but complex — agents need context to query effectively
- ⚠ Official MCP from acryldata — high quality maintenance
- ⚠ Large enterprises may have millions of data assets — pagination critical
- ⚠ GraphQL API provides richer queries than REST — check which the MCP uses
- ⚠ Token expiration policies vary by DataHub version
Alternatives
Full Evaluation Report
Detailed scoring breakdown, competitive positioning, security analysis, and improvement recommendations for DataHub MCP Server.
Scores are editorial opinions as of 2026-03-06.