live wire
IBM makes watsonx Orchestrate AgentOps, custom LLM judging and Bedrock-agent discovery generally availableIBMSchemaGate 0.1.45 fixes broken Oracle ADB wallet connections and an OCI stack pinned 28 releases behindSchemaGatePDI’s Amazon Quick procurement agent grounds spend answers in vendor, category and contract contextAWS Business Intelligence BlogBigQuery’s ML.METRICS example returns 0.84 accuracy but 0.30 macro-F1 on the same 100-row classification queryGoogle Cloud BigQuery docsSchemaGate 0.1.44 auto-selects sentence embeddings, lifting bundled-schema retrieval from 90/98 to 93/98SchemaGateSchemaGate 0.1.43 adds read-only SQL execution with per-principal table checks—and documents unauthenticated client assertionsSchemaGateDatabox adds reusable AI Analyst Skills with personal/company scope, auto-matching and marketplace installsDataboxFabric previews an AI builder for data-agent instructions, source guidance and example queriesMicrosoft FabricDatabricks trains data-agent retriever to stop early or spend bounded extra search steps, reporting 5.8-second latencyDatabricksThoughtSpot adds SpotterCode coding agent to its Visual Embed PlaygroundThoughtSpotLongMemEval-S audit: 67–73% of restore-fixable 80k-budget errors came from evicted evidence under three policiesarXivSchemaGate 0.1.42 adds dimension-aware retrieval and fixes complex multi-table SQL promptsSchemaGateSnowflake agent toolsets can silently drop inherited tools when callers lack accessSnowflake DocumentationLooker’s VS Code extension reaches GA with MCP-assisted LookML generation, editing and validationGoogle Cloud Looker release docsIBM makes watsonx Orchestrate AgentOps, custom LLM judging and Bedrock-agent discovery generally availableIBMSchemaGate 0.1.45 fixes broken Oracle ADB wallet connections and an OCI stack pinned 28 releases behindSchemaGatePDI’s Amazon Quick procurement agent grounds spend answers in vendor, category and contract contextAWS Business Intelligence BlogBigQuery’s ML.METRICS example returns 0.84 accuracy but 0.30 macro-F1 on the same 100-row classification queryGoogle Cloud BigQuery docsSchemaGate 0.1.44 auto-selects sentence embeddings, lifting bundled-schema retrieval from 90/98 to 93/98SchemaGateSchemaGate 0.1.43 adds read-only SQL execution with per-principal table checks—and documents unauthenticated client assertionsSchemaGateDatabox adds reusable AI Analyst Skills with personal/company scope, auto-matching and marketplace installsDataboxFabric previews an AI builder for data-agent instructions, source guidance and example queriesMicrosoft FabricDatabricks trains data-agent retriever to stop early or spend bounded extra search steps, reporting 5.8-second latencyDatabricksThoughtSpot adds SpotterCode coding agent to its Visual Embed PlaygroundThoughtSpotLongMemEval-S audit: 67–73% of restore-fixable 80k-budget errors came from evicted evidence under three policiesarXivSchemaGate 0.1.42 adds dimension-aware retrieval and fixes complex multi-table SQL promptsSchemaGateSnowflake agent toolsets can silently drop inherited tools when callers lack accessSnowflake DocumentationLooker’s VS Code extension reaches GA with MCP-assisted LookML generation, editing and validationGoogle Cloud Looker release docs
nl2sql.ai
guideGUIDE

Fabric’s query-to-action path has one deceptively simple trigger: any returned row

Microsoft’s accessible Activator documentation shows how a Warehouse SELECT becomes an operational rule—and why result shape, frequency and throttling need explicit tests.

Fabric query returns rows and triggers an action; zero rows means no trigger.
AI-generated illustration
By The News Desk· Sep 9, 2026the quick take — two AI hosts go live when you do

Microsoft Fabric can turn a Warehouse SQL query into a recurring operational rule, but the trigger contract is narrower than the phrase “alert on a query” suggests. The documented rule fires when the query returns results; if the query returns no value, no alert is raised. The query must also be a SELECT statement. That makes row existence—not a hidden comparison inside Activator—the first control surface for teams wiring generated SQL into actions.

Encode the condition in SQL

The setup starts in the Warehouse SQL query editor: run the query, choose Create rule, select how frequently Fabric should rerun it, and keep the condition set to evaluate on each event. The practical implication is that the safest query is one whose result set already represents the exception. A threshold check such as “show orders breaching the SLA” should return only breaching rows and return nothing when the system is healthy.

That design deserves three tests before activation: a known-positive case that returns the intended row, a known-negative case that returns zero rows, and a repeat run while the underlying condition remains true. Microsoft’s broader Activator documentation says rule previews can estimate historical firing frequency and that state transitions are used to suppress noise, but teams should still verify the behavior of their specific Warehouse polling rule rather than assume streaming-rule semantics carry over unchanged.

Treat the action as production code

A matching query can send email or Teams notifications, invoke a Power Automate custom action, or start Fabric work including a pipeline, dataflow, notebook, Spark job, function or copy job. Parameters can be passed to several Fabric item types, while copy jobs do not accept them. Microsoft also documents a Test action control before a rule is started.

The limits matter. Activator allows up to 500 rules per item and 10,000 incoming events per second per rule. Action ceilings include 500 email or Teams messages per Activator item per hour, 10,000 Power Automate flow executions per rule per hour, and 50 Fabric-item activations per user per minute. Exceeding an action limit can lead to throttling or cancellation.

The deployment checklist

For analytics-agent teams, the operational boundary is clear: do not let free-form generated SQL flow directly into an enabled action. Review and save a bounded SELECT; verify positive, negative and persistent-condition cases; set an explicit polling frequency; test the downstream action with representative parameters; then monitor it against the documented quotas.

Fabric’s convenience is real: a validated warehouse query can become a running workflow without a separate scheduler. The same shortcut also collapses query logic and operational triggering into one artifact. That is useful only when the SQL’s zero-row and nonzero-row behavior is treated as an API contract, not as an incidental query result.

sources

Filed by The News Desk. Corrections: desk@nl2sql.ai · Our standards →

comments · 0

    Comments are moderated before they appear. Your email is used once to confirm it is you — never shown, never sold. Corrections and questions get an answer from the desk when we have one.