datahub-project/datahub

Health v2 · maintenance only

Adoption not part of the score

12586 stars · 3675 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

DataHub is an open-source data catalog and metadata platform providing data discovery, governance, lineage, and observability across an organization's data and AI stack. It ingests metadata from 150+ integrations and exposes it via UI, APIs, SDKs, and MCP-based context services for AI agents.

Use cases

• catalog all datasets, dashboards, and ML models in one searchable place
• track column-level data lineage across pipelines
• set up data quality checks and anomaly detection
• assign data ownership and tag PII for governance
• give AI agents trusted metadata context via MCP
• self-host a data catalog on Kubernetes or Docker

When to choose

• you need an enterprise-grade, self-hosted data catalog with broad connector support
• you want unified discovery, lineage, and observability across a large data estate
• you need programmatic metadata management via APIs and SDKs

When to avoid

• you only need lightweight documentation of a handful of tables
• you want a simple spreadsheet-style data inventory without infrastructure
• your team cannot operate a multi-service Docker/Kubernetes deployment

Facets

application · maturity active

search-engine monitoring etl api-framework mcp analytics databases analytics data-science large-language-models self-hosted developer-tools python self-hosted cloud data-catalog metadata-management data-governance data-discovery data-lineage data-observability data-quality context-management metadata-ingestion data-contracts data-engineering docker kubernetes web-server

Member repositories

For agents

markdown · JSON · MCP: product_card(name="datahub-project/datahub")