live wire
▸Oracle Select AI generates SQL/PGQ for property graphs—but errors if one AI profile mixes graphs with tables or viewsOracle▸Domo’s data team standardizes MCP builds with shared scaffolding; beta toolkits inherit user permissionsDomo▸Databricks proposes five outcome signals for agentic BI—and says dashboard counts and license usage miss decision qualityDatabricks▸One vibe-coded database app needed 56+ human test hours and three repair rounds despite AI Studio’s verification claimsMancas & Mancas / arXiv▸BigQuery AI.IF’s default cost mode trains a local model for most rows—and drops row-error thresholdsGoogle Cloud▸AWS’s multi-tenant data-agent pattern gives the broker Lambda no data permissions; Cognito roles carry Lake Formation scopeAWS Big Data Blog▸Amazon OpenSearch’s optimized engine adds nested agent traces, but AWS reserves it for log analytics—not vector searchAWS▸Two interpretation-based provenance semantics cannot express SQL bag difference, new theorem showsKolaitis / arXiv▸Databricks Feature Views reuse one definition for training and live scoring, but online lookup still returns latest available stateDatabricks▸Databricks Genie’s Slack and Teams previews default channel answers to public, even when viewers lack source-data accessDatabricks▸Snowflake’s two-tool GraphRAG agent beat a seven-tool graph setup 70% to 60% on 22 ontology questionsSnowflake Engineering▸Semantix opens an Apache-2.0 semantic-first data-agent stack, but labels the project WIP and has no release or adoption yetGitHub / icosmos-space▸Confluent defines Streamhouse as shared current state for agents and apps, but says broader semantics and governance are still to comeConfluent▸Graph-repair retrieval cut F1 by 0.126 on one suite; a learned router used fewer tokens than always-on retrieval in all sixLin et al. / arXiv▸Oracle Select AI generates SQL/PGQ for property graphs—but errors if one AI profile mixes graphs with tables or viewsOracle▸Domo’s data team standardizes MCP builds with shared scaffolding; beta toolkits inherit user permissionsDomo▸Databricks proposes five outcome signals for agentic BI—and says dashboard counts and license usage miss decision qualityDatabricks▸One vibe-coded database app needed 56+ human test hours and three repair rounds despite AI Studio’s verification claimsMancas & Mancas / arXiv▸BigQuery AI.IF’s default cost mode trains a local model for most rows—and drops row-error thresholdsGoogle Cloud▸AWS’s multi-tenant data-agent pattern gives the broker Lambda no data permissions; Cognito roles carry Lake Formation scopeAWS Big Data Blog▸Amazon OpenSearch’s optimized engine adds nested agent traces, but AWS reserves it for log analytics—not vector searchAWS▸Two interpretation-based provenance semantics cannot express SQL bag difference, new theorem showsKolaitis / arXiv▸Databricks Feature Views reuse one definition for training and live scoring, but online lookup still returns latest available stateDatabricks▸Databricks Genie’s Slack and Teams previews default channel answers to public, even when viewers lack source-data accessDatabricks▸Snowflake’s two-tool GraphRAG agent beat a seven-tool graph setup 70% to 60% on 22 ontology questionsSnowflake Engineering▸Semantix opens an Apache-2.0 semantic-first data-agent stack, but labels the project WIP and has no release or adoption yetGitHub / icosmos-space▸Confluent defines Streamhouse as shared current state for agents and apps, but says broader semantics and governance are still to comeConfluent▸Graph-repair retrieval cut F1 by 0.126 on one suite; a learned router used fewer tokens than always-on retrieval in all sixLin et al. / arXiv
nl2sql.ai
guideDeployment guide

Tableau’s MCP-ready knowledge graph is available now—but only inside its premium cloud tiers

Tableau Knowledge can ground external agents in governed analytics assets, yet its deployment, licensing and connector boundaries matter before teams treat it as a universal semantic layer.

By The News Desk· Nov 7, 2026the quick take — two AI hosts go live when you do

Tableau has started selling Tableau Knowledge as a governed context layer for analytics agents. The product turns published data sources, workbooks, dashboards and metrics into knowledge graphs that Tableau Agent—or an external MCP-enabled agent—can retrieve from. That makes it relevant to teams building natural-language analytics, but the launch is narrower than the phrase “any AI agent” may imply. Tableau’s product announcement says the service is available now in Tableau Cloud+ and Tableau+, while Tableau Server, Tableau Next and connectors to several outside platforms are still described as coming soon.

What is available now

The current product harvests Tableau assets, infers relationships and presents a no-code interface where subject-matter experts can inspect, curate and enrich the graph. Tableau says source changes flow back into the graph, conflicts and duplicates are surfaced, and downstream agent feedback can refine it. The graph is accessible to Tableau Agent and via MCP, allowing other compatible agents to retrieve governed context rather than search every data source independently. Tableau’s product page also lists published data sources, workbooks, dashboards, catalog content and unstructured material such as PDFs and wikis as inputs.

For an analytics engineering team, that creates two separate control planes to review. Tableau governs the knowledge that an agent can retrieve; the calling agent still governs what happens after retrieval. An MCP connection is an access path, not proof that a generated query, chart or action is authorized. Teams should therefore test identity propagation, content permissions, logging and write capabilities at the external-agent boundary rather than assuming the graph’s governance automatically covers the entire workflow.

The edition boundary is part of the architecture

Tableau’s pricing page puts Tableau Knowledge in Cloud+, which also includes Tableau Agent, and in the Tableau+ bundle. Both require contacting sales; Tableau does not publish a list price for those tiers. Standard starts at $15 per user per month and Enterprise at $35 per user per month when billed annually, but neither edition lists Tableau Knowledge. Tableau also says all products require annual contracts.

That packaging matters for pilots. A team cannot validate the production feature by giving a few users ordinary Standard or Enterprise seats. Tableau’s role descriptions say Creators can create and edit Knowledge graphs, Explorers can interact with them, and Viewers can view and interact with them, but the feature still sits behind the premium edition boundary.

Questions to settle before rollout

Start with the sources the product can ingest today, not the future roadmap. Tableau’s announcement says Snowflake, Databricks, Workday and other external-source connections are coming soon. Confirm whether the assets needed for the first use case already live in Tableau and define who can approve inferred relationships or resolve conflicting definitions.

Then test the MCP path as its own integration: which graph objects are exposed, how Tableau permissions are represented to the client, what is logged, and whether feedback from agent interactions can change future retrieval. Finally, price the deployment using the actual Cloud+ or Tableau+ contract, including the mix of graph authors and consumers. The product can reduce context fragmentation, but only after teams account for the premium-edition gate and the security boundary beyond Tableau’s graph.

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