generated: '2026-09-05' method: derived source: json-schema/databricks-asset-bundles-bundle-jsonschema.json (first-party, fetched 2026-09-05 from https://raw.githubusercontent.com/databricks/cli/main/bundle/schema/jsonschema.json) note: Derived mechanically from the Databricks-published bundle JSON Schema. Entity names and the resource list are read out of the schema, not written by hand; no relationship is asserted that the schema does not express. root: databricks.yml entities: - name: Bundle path: bundle description: Core bundle identity — name, uuid, databricks_cli_version constraint, git origin, deployment settings. relationships: - kind: has_many to: Target via: targets - kind: has_many to: Resource via: resources - kind: has_many to: Variable via: variables - kind: has_one to: Workspace via: workspace - name: Workspace path: workspace description: Workspace connection and path defaults — host, profile, auth_type, root_path. relationships: - kind: belongs_to to: Bundle via: workspace - name: Target path: targets description: A named environment (dev/staging/prod) whose overrides replace workspace, resource, preset and run_as settings for one deployment. relationships: - kind: belongs_to to: Bundle via: targets - kind: has_one to: Workspace via: targets..workspace - kind: has_many to: Resource via: targets..resources - kind: has_one to: Presets via: targets..presets - name: Resource path: resources description: A Databricks object the bundle owns. The schema defines 34 resource collections. resource_types: - alerts - apps - catalogs - cluster_policies - clusters - dashboards - database_catalogs - database_instances - experiments - external_locations - genie_spaces - instance_pools - job_runs - jobs - model_serving_endpoints - models - pipelines - postgres_branches - postgres_catalogs - postgres_databases - postgres_endpoints - postgres_projects - postgres_roles - postgres_synced_tables - quality_monitors - registered_models - schemas - secret_scopes - secrets - sql_warehouses - synced_database_tables - vector_search_endpoints - vector_search_indexes - volumes relationships: - kind: belongs_to to: Bundle via: resources - kind: has_many to: Permission via: resources...permissions - name: Variable path: variables description: A typed named value referenced elsewhere in the bundle; may resolve by lookup against a workspace object. relationships: - kind: belongs_to to: Bundle via: variables - name: Artifact path: artifacts description: A build output (Python wheel, JAR) built and uploaded by the CLI during deploy. relationships: - kind: belongs_to to: Bundle via: artifacts - kind: has_many to: Resource via: library reference - name: Permission path: permissions description: An access grant applied to bundle-declared resources. relationships: - kind: belongs_to to: Bundle via: permissions - name: Presets path: presets description: Cross-cutting defaults (name prefix, tags, pause status, concurrency) applied to every resource in a target. relationships: - kind: belongs_to to: Target via: presets - name: Sync path: sync description: File inclusion/exclusion and sync paths for what is uploaded to the workspace. relationships: - kind: belongs_to to: Bundle via: sync - name: Scripts path: scripts description: Named scripts runnable as bundle hooks. relationships: - kind: belongs_to to: Bundle via: scripts top_level_mappings: - artifacts - bundle - environments - experimental - include - permissions - presets - python - resources - run_as - scripts - sync - targets - variables - workspace resource_type_count: 34 id_prefixes: null gaps: - Databricks publishes no OpenAPI, so no operation-level entity graph could be derived; this graph describes the configuration model the schema defines, which is the contract this product actually ships.