Unity Catalog

Open-source universal data and AI catalog from Databricks providing unified governance for tables, files, models, and functions across clouds and compute engines. Unity Catalog (the open-source version) enables organizations to register and govern data assets with fine-grained access control, data lineage, and a unified namespace outside of Databricks. Supports Delta Lake tables, volumes (files), registered ML models, and SQL functions. The open-source release aims to make Unity Catalog the default catalog standard for data ecosystems.

Evaluated Mar 06, 2026 (0d ago) v0.2+
Homepage ↗ Repo ↗ Other data-catalog governance metadata open-source databricks delta-lake tables models functions
⚙ Agent Friendliness
56
/ 100
Can an agent use this?
🔒 Security
84
/ 100
Is it safe for agents?
⚡ Reliability
63
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
--
Documentation
72
Error Messages
70
Auth Simplicity
80
Rate Limits
82

🔒 Security

TLS Enforcement
90
Auth Strength
82
Scope Granularity
85
Dep. Hygiene
82
Secret Handling
80

Apache 2.0, Databricks-backed. Fine-grained RBAC at catalog/schema/table/column level. OAuth2 auth. Early-stage open source — security posture improving. Databricks managed version has enterprise security track record.

⚡ Reliability

Uptime/SLA
68
Version Stability
60
Breaking Changes
55
Error Recovery
70
AF Security Reliability

Best When

You're building a multi-engine, multi-cloud data platform and want a vendor-neutral open-source catalog that works with Spark, Trino, DuckDB, and other engines.

Avoid When

You're in a pure Databricks environment (use managed Unity Catalog) or need a mature, battle-tested catalog (use Apache Atlas or Collibra).

Use Cases

  • Create a unified catalog for tables, ML models, and files across multi-cloud data platforms with consistent access control
  • Register Delta Lake tables and volumes from multiple compute engines (Spark, Trino, DuckDB) under a single namespace
  • Govern AI models alongside their training data in one catalog — trace which model was trained on which dataset version
  • Provide agents with a catalog API to discover available datasets, tables, and models and their schemas without knowing the underlying storage
  • Implement column-level security and row-level filtering across data assets using Unity Catalog's fine-grained access controls

Not For

  • Teams using Hive Metastore exclusively — Unity Catalog is a newer standard that may require migration effort
  • Simple single-engine analytics — Hive Metastore or AWS Glue are simpler if you only use one compute engine
  • Production deployments today — open source Unity Catalog is in early stages; Databricks' managed version is production-ready

Interface

REST API
Yes
GraphQL
No
gRPC
No
MCP Server
No
SDK
Yes
Webhooks
No

Authentication

Methods: bearer_token
OAuth: Yes Scopes: Yes

OAuth2/JWT for REST API authentication. Databricks managed Unity Catalog uses Databricks PAT tokens. Open source: configurable auth. RBAC at catalog, schema, and object level (GRANT/REVOKE SQL syntax).

Pricing

Model: open_source
Free tier: Yes
Requires CC: No

Apache 2.0 licensed open source. Databricks manages Unity Catalog as part of their commercial platform. Open source is for self-hosting outside Databricks.

Agent Metadata

Pagination
cursor
Idempotent
Full
Retry Guidance
Not documented

Known Gotchas

  • Open source Unity Catalog is in active development (v0.x) — expect breaking changes between releases
  • Databricks' managed Unity Catalog and open source Unity Catalog have different feature sets — verify which version documentation applies
  • Delta sharing protocol integration requires Delta Sharing server configuration — Unity Catalog alone doesn't serve data files
  • Column-level security (column masks, row filters) requires specific compute engine support — not all engines enforce UC policies
  • Three-level namespace (catalog.schema.table) is mandatory — agents must qualify all object references with full namespace path
  • MCP server for Unity Catalog exists but is early stage — verify stability before production agent workflows
  • Credential vending for underlying storage requires cloud IAM configuration — Unity Catalog doesn't manage cloud credentials by default

Alternatives

Full Evaluation Report

Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for Unity Catalog.

AI-powered analysis · PDF + markdown · Delivered within 30 minutes

$99

Package Brief

Quick verdict, integration guide, cost projections, gotchas with workarounds, and alternatives comparison.

Delivered within 10 minutes

$3

Score Monitoring

Get alerted when this package's AF, security, or reliability scores change significantly. Stay ahead of regressions.

Continuous monitoring

$3/mo

Scores are editorial opinions as of 2026-03-06.

5382
Packages Evaluated
26151
Need Evaluation
173
Need Re-evaluation
Community Powered