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

Oracle’s Gemini integration keeps the data agent inside the database

Gemini becomes the chat client; the Select AI agent team remains installed in Autonomous AI Database and is reached over A2A with database client credentials.

Gemini chats with an in-database Oracle AI agent through A2A, not outside it.
Side by side: what changed
By The News Desk· Sep 9, 2026the quick take — two AI hosts go live when you do

Oracle has documented a new path for using its in-database AI agent teams from Google Gemini Enterprise. The important architectural choice is not the chat interface; it is where the database reasoning continues to run.

Gemini is the client, not the database agent

Oracle says the integration lets A2A client applications call in-database AI agents and extends that route to Google Gemini Marketplace. Users can query a Select AI agent team through the Gemini Enterprise chat interface, but Oracle’s setup still requires the agent team to be installed inside Autonomous AI Database. In other words, Gemini is the conversational entry point while the configured database agent remains the system that handles the data-side task. Oracle’s integration guide

That distinction matters for teams evaluating cross-vendor data-agent integrations. A broad assistant surface does not have to absorb the database-specific agent implementation. Oracle’s documented contract instead uses A2A to connect the client agent to an existing in-database team. The interface can change without moving the team itself out of the database. Oracle’s integration guide

The credential boundary remains explicit

The setup is not credential-free. Oracle lists two prerequisites: the in-database AI agent team must already be installed, and administrators must create client credentials for Autonomous AI Database. That makes the external chat surface only one part of the deployment. Operators still have to provision an agent team and an authenticated client path into the database. Oracle’s integration guide

Oracle’s September AI roundup describes the broader direction as natural-language access to database agents from external enterprise AI experiences. The same roundup separately highlights OCI Enterprise AI NL2SQL controls such as selectable models, scheduled semantic-store enrichment and background SQL generation. Those are separate capabilities, but together they show Oracle treating conversational SQL as an operated service rather than a prompt pasted into a general chatbot. Oracle’s September AI roundup

What the documentation does not establish

The short guide does not publish accuracy, latency or production-scale results for the Gemini path. It also does not say that Gemini receives unrestricted database access. The defensible reading is narrower: Gemini Enterprise can act as an A2A client for an installed Select AI agent team, and Autonomous AI Database client credentials remain a prerequisite.

For practitioners, the review checklist is therefore concrete: inspect the in-database team, scope the client credentials and test the A2A boundary. The product news is cross-vendor chat access; the deployment work is still database-side governance.

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.

    more from the desk