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
releaseMICROSOFT FABRIC

Microsoft gives Fabric data agents a native charting layer

The preview moves natural-language answers onto Fabric’s shared visualization foundation, with automatic charts across lakehouses, warehouses and Power BI semantic models.

Fabric data agents switch to shared charts
AI-generated illustration
By The News Desk· Aug 22, 2026the quick take — two AI hosts go live when you do

Microsoft is upgrading the presentation layer of Fabric data agents, replacing their standalone chart renderer with Fabric Visuals, the same visualization foundation used by Fabric Apps and increasingly elsewhere in the platform. The change is in preview and applies to the visual responses that accompany answers to natural-language questions. Microsoft says the aim is more consistent formatting, legends, tooltips, markers and axis scaling across Fabric experiences.

What shipped

The data agent can now return line and multi-line charts, bar and multi-bar charts, stacked bars, pie charts, scatter plots, and area or stacked-area charts using the shared Fabric renderer. Microsoft says visual responses work across every source type currently supported by the agent: lakehouses, warehouses and Power BI semantic models.

No administrator switch is required. The agent decides when a visualization would help, including when a user asks for a chart, uses language such as “show me,” or poses a question the system judges clearer as a visual. Users can also name a preferred chart type in the prompt, while agent instructions can establish preferences for particular classes of questions. Microsoft says visuals are enabled by default in data-agent conversations.

Why it matters for NL2SQL teams

This is not a new text-to-SQL engine; it is a product-layer change after the query and answer are produced. But that layer matters in enterprise deployments. A conversational analytics tool that returns a correct table but makes users rebuild the result in a dashboard still leaves a workflow gap. Microsoft is trying to keep the question, result and interpretation in one surface while making the output behave more like the rest of Fabric. The company’s examples show natural-language requests producing multi-series line, clustered bar and stacked-area charts directly in the conversation.

The practical boundary remains important: a polished chart does not by itself establish that the underlying query, metric definition or selected source was correct. Teams evaluating the preview should therefore test visual rendering separately from semantic accuracy—especially axis choices, category ordering and automatic chart selection—while retaining their existing validation for generated answers. Microsoft positions the shared renderer as a foundation for richer tooltips, interactivity and accessibility improvements to come, but those additions are not part of the current announcement. The preview announcement describes rendering and supported chart types rather than changes to query generation or semantic validation.

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.