Unifying enterprise context and semantics: LookML in Knowledge Catalog is now GA

Google Cloud has announced the general availability of LookML semantic language indexing in Knowledge Catalog for Looker Core on Google Cloud. This integration addresses a critical gap where AI agents previously struggled to distinguish between data access and data understanding. By combining organizational context from Knowledge Catalog with deterministic business semantics from LookML, enterprises can now provide AI agents with both broad metadata and exact metric definitions, join paths, and SQL formulas.

The integration ensures that LookML’s business logic, which powers trusted BI reporting, is automatically indexed into Knowledge Catalog. This enables AI agents like Gemini Enterprise to retrieve enterprise-wide discovery alongside precise metric calculations without manual intervention. Previously, catalog integrations only captured shallow pointers, but now LookML’s deterministic semantics become a first-class layer in the Knowledge Catalog context graph, ensuring agents operate on the freshest business logic in near-real-time.

With this release, agents no longer need to infer mathematical relationships or guess SQL logic, reducing errors and token waste. Knowledge Catalog enforces governance while LookML provides the exact SQL definitions and join conditions. Agents can now return accurate answers alongside interactive links to Looker Explores, allowing users to inspect data directly. Synchronization is automatic, eliminating the need for custom agent code or separate data stores.

To implement this, enterprises can connect existing LookML business logic to enterprise AI assistants like Gemini Enterprise with minimal setup. Google Cloud provides documentation for managing Looker Core with Knowledge Catalog and guides for connecting to Knowledge Catalog in Gemini Enterprise. Developers can also explore customization options through an ADK agent codelab on GitHub.