Classification and Masking

The current repository includes starter taxonomy and masking configuration under governance/.

Taxonomy model

Taxonomy files define categories such as:

• internal
• confidential
• pii_personal

These categories provide structured metadata for how data should be handled.

Masking model

The alpha ships starter built-in masking definitions, including:

• partial_mask
• full_mask

These are currently compiled into internal mask resources so the rest of the framework can reason about them consistently.

Current boundary

Taxonomy and masking definitions are validated and compiled into deterministic resources. The OSS alpha does not install live Databricks column masks. Treat these files as versioned governance intent until provider-backed enforcement is implemented.

Roadmap enforcement path

Area	Status	Current guidance
Classification labels	Experimental	Use as reviewed metadata in configuration and plans.
Built-in mask definitions	Experimental	Compile locally for future policy wiring.
Live Databricks column masks	Roadmap	Do not rely on DataMuru to enforce masks in 0.5.1a0.
Enterprise policy workflows	Enterprise	Expected to require approval, evidence, and account controls.

For regulated data, keep existing Databricks policies and manual review gates in place. DataMuru can version the intended governance model today, but it should not be the only enforcement control for masking.