Google Cloud announced the general availability of LookML semantic language indexing in Knowledge Catalog for Looker Core on October 26, 2026. The integration bridges the gap between broad enterprise context and deterministic business semantics for AI agents.
What changed
Google Cloud announced on October 26, 2026, the general availability of LookML semantic language indexing in Knowledge Catalog for Looker Core. According to outbound product managers Maire Newton and Ramanan Balakrishnan, the release natively integrates LookML business logic into the Knowledge Catalog context graph.
The synchronization is active by default, ensuring that updates pushed to production LookML appear in Knowledge Catalog in near-real-time. This provides Gemini Enterprise and custom AI agents with both broad organizational context and deterministic BI semantics.
Technical details and schema mapping
Knowledge Catalog captures and links multiple entity levels across Looker Core, including LookML models, Explores, Views, fields, dashboards, and visualization tiles. LookML Explores index relational join logic, filter guardrails, and cardinality rules as Schema-Join relationships.
Views and fields extract compiled SQL expressions, primary keys, and derived table transformation logic. Furthermore, users can view end-to-end lineage from underlying BigQuery tables to individual dashboard tiles.
Integration and access
Metadata is indexed directly into Knowledge Catalog by default, making it available to Gemini Enterprise without custom agent code. When connected, Gemini Enterprise uses read-only skills and hybrid search to query technical schemas, business glossaries, and LookML logic.
Developers building custom agents can access the context bundles using the LookupContext API. The system uses end-to-end user credentials so queries respect existing Identity and Access Management permissions.
What to do
- Confirm first-party metadata indexing is enabled for Looker Core on Google Cloud.
- Enable the Cloud Dataplex API and Data Lineage API in your project.
- Grant users required view roles, including roles/dataplex.viewer, roles/looker.schemaViewer, and roles/datalineage.viewer.
- Enable Knowledge Catalog in Gemini Enterprise by attaching a federated data store or by opening your Gemini Enterprise app in the Google Cloud console, navigating to Configurations and Assistant, toggling on Enable Knowledge Catalog, and clicking Save and publish.
Key facts and where they come from
- Google Cloud announced the general availability of LookML semantic language indexing in Knowledge Catalog for Looker Core on October 26, 2026.
today, we are announcing the general availability (GA) of LookML semantic language indexing in Knowledge Catalog for Looker Core on Google Cloud.
- Synchronization between production LookML and Knowledge Catalog operates live by default.
any update pushed to production LookML is reflected in Knowledge Catalog in near-real-time so agents operate on the freshest business logic.
- Knowledge Catalog indexes entity levels including LookML models, Explores, Views, fields, and dashboards.
Knowledge Catalog now captures and links the following entity levels across Looker Core on Google Cloud:
