Connectors

Bring the file your tools already produce. Each connector documents five paths: get it working, drive it from the API, drive it from an agent, automate it in CI, then go deep.

Looking for why a connector matters rather than how to run it? The guides live on coremodels.io.

Format

One governed reading of the schema that currently lives in four places.

Stop asking a validator four questions it was only built to answer one of.

5 articles · 6 recipes

Make the vocabulary you already trust executable downstream.

Let the pipeline read the model that is already right.

Stop shipping a recollection of the standard as if it were the standard.

Make the contract a derived artifact of governed meaning, not a YAML twin of the table.

Keep the ontology and the application schema from becoming two portraits of Person.

Protocol Buffers

Carry reserved field numbers with the meaning, not only with the .proto file.

Keep the four assertions in every shape constraint from dying in a hand copy.

Give the column nobody can explain a rule the database can actually carry.

Platform

Govern the catalog every sync trusts - before a silent type change lands in the warehouse.

5 articles · 4 recipes

Govern the scheduler that already knows your freshness, ownership, and lineage.

5 articles · 4 recipes

Stop treating a crawler snapshot as the meaning of your lake.

5 articles · 5 recipes

Record the constraints that die on the way to a registered schema.

Govern the four comment rows that are the schema of your study.

5 articles · 5 recipes

Put meaning under every Kafka subject the Schema Registry already inventories.

5 articles · 5 recipes

Give Unity Catalog something to check thousands of columns against.

5 articles · 5 recipes

Meaning has no home in dbt. Bring a manifest; we turn `accepted_values` into a vocabulary you can govern and your agent can read.

5 articles · 8 recipes · 7 videos

Google BigQuery

Turn INFORMATION_SCHEMA from an inventory into a contract you can audit.

5 articles · 5 recipes

Microsoft Fabric

Catch the warehouse type change that never pages anyone.

5 articles · 5 recipes

Write down the graph schema MERGE has been inferring for you.

5 articles · 5 recipes

Open Semantic Interchange

Put two revenue numbers next to the same definition.

Diff the codebook so a study change is a review, not a memory.

5 articles · 4 recipes

Govern the org whose schema can change in forty seconds from Setup.

5 articles · 5 recipes

Your Snowflake schema is not a contract. Generate one from the governed model.

5 articles · 4 recipes