@fabric-harness/databricks 4.0.2 → 4.1.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/README.md CHANGED
@@ -89,7 +89,7 @@ services.
89
89
 
90
90
  - AI Gateway and Model Serving: automatic `system.ai.*` routing, explicit mode/base URL selection, model-service discovery, gateway readiness, request tags, streaming, tools, and custom endpoints.
91
91
  - Identity: Databricks CLI OAuth profiles, explicit PAT/pre-fetched bearer, cached OAuth M2M, Databricks Apps service principal, and forwarded user token helpers.
92
- - Data and AI: SQL Warehouse, Unity Catalog discovery/admin, AI Search query/admin, bounded citation-validated RAG with token streaming, MLflow 3 evaluation records and a managed-judge release Job, embeddings, `ai_query()`, typed Genie Agent conversations, beta normalized lifecycle/ACL authoring, explicit Beta Agent Mode SSE streaming, Feature Serving, typed Jobs 2.2 authoring/run APIs, and workspace read/write APIs.
92
+ - Data and AI: SELECT-only `databricksSqlReadTool()`, approval-bound arbitrary SQL Warehouse execution, Unity Catalog discovery/admin, AI Search query/admin, bounded citation-validated RAG with token streaming, MLflow 3 evaluation records and a managed-judge release Job, embeddings, `ai_query()`, typed Genie Agent conversations, beta normalized lifecycle/ACL authoring, explicit Beta Agent Mode SSE streaming, Feature Serving, typed Jobs 2.2 authoring/run APIs, and workspace read/write APIs.
93
93
  - Agent governance: explicit-Beta Unity Catalog Agent Services registration, discovery, metadata updates, `EXECUTE`/`READ_METADATA` grants, and deletion for externally hosted Harness agents.
94
94
  - Data engineering: Lakeflow pipeline list/status/start/stop plus separately opt-in create/update/delete/event tools.
95
95
  - Operations: System Tables consumption reporting, actual-cost budget source, MLflow tracing, serving usage capture, and Lakebase telemetry tables.
@@ -114,6 +114,12 @@ resource descriptors validate at bundle initialization and check every catalog-q
114
114
  Request-scoped `bundle.forPrincipal({ tokenProvider, principal })` uses a verified principal and
115
115
  binds approval grants to the exact call, input digest, and executing identity. Secret write tools
116
116
  accept only SDK `SecretRef` values resolved by a server-side `SecretProvider`.
117
+
118
+ For an analytics copilot, `analyticsCopilotGovernance()` keeps ordinary Genie questions and
119
+ SELECT-only `sql_read` calls approval-free while arbitrary SQL and Genie management stay routed to
120
+ the steward audience. Its SQL/Genie service scope does not weaken authoring validation: enabling
121
+ another authoring service fails initialization until approval routing explicitly includes it.
122
+
117
123
  Existing clusters require an explicit `allowedExistingClusterIds` entry, and model-facing serving
118
124
  tools reject raw `environmentVars`.
119
125
  Genie management is enabled by `genie.manage` and requires a `DatabricksManagedResourceStore`; use
package/dist/agent.js CHANGED
@@ -19977,7 +19977,7 @@ function databricksGenieTool(client, options) {
19977
19977
  properties: { question: { type: "string" }, conversationId: { type: "string" } },
19978
19978
  required: ["question"]
19979
19979
  },
19980
- metadata: { effect: "execute", provider: "databricks", service: "genie" },
19980
+ metadata: { effect: "read", provider: "databricks", service: "genie" },
19981
19981
  execute(input, context) {
19982
19982
  return genie.ask({
19983
19983
  question: input.question,