<!-- mobian-agent-page publisher="dailydev" canonical="https://daily.dev/posts/microsoft-fabric-becomes-a-context-layer-for-ai-agents-with-clickhouse-added-for-fast-analytics-of5kmyizq" -->

---
title: Microsoft Fabric becomes a context layer for AI agents,...
description: Microsoft used FabCon and SQLCon in Barcelona to reposition Fabric as a context layer for enterprise AI agents. Fabric IQ, its context layer, now feeds...
canonical: https://daily.dev/posts/microsoft-fabric-becomes-a-context-layer-for-ai-agents-with-clickhouse-added-for-fast-analytics-of5kmyizq
twitter:card: summary_large_image
twitter:site: @dailydotdev
og:type: website
og:site_name: daily.dev
og:title: Microsoft Fabric becomes a context layer for AI agents, with ClickHouse added for fast analytics | daily.dev
og:description: Microsoft used FabCon and SQLCon in Barcelona to reposition Fabric as a context layer for enterprise AI agents. Fabric IQ, its context layer, now feeds...
og:url: https://daily.dev/posts/microsoft-fabric-becomes-a-context-layer-for-ai-agents-with-clickhouse-added-for-fast-analytics-of5kmyizq
og:image: https://api.daily.dev/og/posts/OF5KMyizQ.png
og:image:alt: Microsoft Fabric becomes a context layer for AI agents, with ClickHouse added for fast analytics
og:image:width: 1200
og:image:height: 630
og:locale: en
---

> ## Documentation Index
> Fetch the complete documentation index at: https://daily.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# Microsoft Fabric becomes a context layer for AI agents, with ClickHouse added for fast analytics

**[Collections](https://daily.dev/sources/collections)** · 3 min read · 2 upvotes · 0 comments

## Summary

Microsoft used FabCon and SQLCon in Barcelona to reposition Fabric as a context layer for enterprise AI agents. Fabric IQ, its context layer, now feeds Microsoft 365 Copilot by default and is queryable by outside agents via MCP (Fabric IQ MCP is generally available with read-only tools). Ontologies, which map business meaning onto semantic models, moved further into preview. Power BI is gaining app creation with write-back capability. Additional announcements include mirrored security roles from Snowflake, two-way Salesforce Data Cloud 360 integration, BigQuery mirroring, and a new Osmos-based data engineering agent usable from GitHub Copilot, VS Code, Codex, and Claude Code. Separately, a native ClickHouse workload for Fabric entered public preview, syncing OneLake tables as accelerated copies on dedicated Azure compute, with governed read-only access for AI agents, Power BI, and notebooks, and planned future support for write-back, replication, permission sync, and vector search.

## Content

At FabCon and SQLCon in Barcelona, Microsoft made a clear bet: the hard part of enterprise AI agents is context, not models. Fabric is now being repositioned as the place where agents learn how a business actually works.

## Fabric IQ and the context layer

Fabric IQ is the platform's context layer. It now feeds Microsoft 365 Copilot by default, and outside agents can query it over MCP. Fabric IQ MCP is generally available with read-only tools.

Ontologies are the piece that does the heavy lifting. They layer business rules and relationships on top of semantic models, so an agent knows what a "customer" or an "active order" means in your company rather than guessing from column names. They moved further into preview at the conference.

Power BI is getting an app-creation feature that turns semantic models into full applications, including write-back. That means a model built for reporting can also let people change data.

## Other Fabric announcements

- Mirrored security roles from sources like Snowflake, so access rules follow the data
- Two-way integration with Salesforce Data Cloud 360
- BigQuery mirroring
- A new data engineering agent built on the acquired Osmos technology, which can be driven from GitHub Copilot, VS Code, Codex, and Claude Code

Microsoft isn't alone here. Databricks, Snowflake, Google, and Salesforce have all shipped comparable context-layer products. Everyone seems to have reached the same conclusion at roughly the same time: an agent with no grounding in business meaning is expensive and unreliable.

## ClickHouse on OneLake, now in public preview

The most concrete addition is a native ClickHouse workload for Fabric, in public preview. It provisions a dedicated ClickHouse Cloud service to speed up analytics on OneLake data.

How it works:

- OneLake tables sync as accelerated copies, with no custom ETL pipelines
- An embedded SQL console is included
- Governed, read-only access is available to AI agents (Copilot Studio, GitHub Copilot, MCP-compatible tools), Power BI, and Fabric notebooks
- It runs on dedicated Azure compute, so agent and dashboard workloads don't compete with Fabric capacity

That last point matters more than it sounds. Agents can generate a lot of queries, and having them share capacity with dashboards is a good way to make both slow.

ClickHouse says it plans to add write-back to OneLake, continuous replication, permission sync, and full-text and vector search.

## The pattern

Taken together, the two announcements fit: Fabric IQ supplies the meaning, and workloads like ClickHouse supply the speed. Whether agents built on this stack are actually better than the ones built on its competitors' stacks is a question the preview period will have to answer.

## Questions this post answers

### What is Fabric IQ and how do AI agents access it?

Fabric IQ is Microsoft Fabric's context layer that feeds Microsoft 365 Copilot by default and exposes data to outside agents via MCP. Fabric IQ MCP is generally available with read-only tools, and ontologies layering business rules onto semantic models moved further into preview, letting agents understand terms like 'customer' or 'active order' correctly.

_Teams grounding AI agents in real business context can follow Fabric IQ's rollout on daily.dev._

### How does the new ClickHouse integration in Microsoft Fabric work?

A native ClickHouse workload for Fabric is in public preview, provisioning a dedicated ClickHouse Cloud service that syncs OneLake tables as accelerated copies without custom ETL. It includes an embedded SQL console, governed read-only access for AI agents (Copilot Studio, GitHub Copilot, MCP tools), Power BI, and notebooks, and runs on dedicated Azure compute separate from Fabric capacity so agent queries don't slow dashboards.

_Anyone speeding up analytics on OneLake can track ClickHouse's preview features on daily.dev._

### Which coding tools can drive Microsoft's new Osmos-based data engineering agent?

The Osmos-based data engineering agent added to Fabric can be driven from GitHub Copilot, VS Code, Codex, and Claude Code, giving developers multiple entry points to automate data engineering tasks directly from familiar coding environments rather than a separate Fabric-only interface.

_Developers wiring coding assistants into data workflows can follow these integrations on daily.dev._

## Similar posts on daily.dev

- [ClickHouse expands collaboration with Microsoft, bringing Fabric integration, deeper OneLake interoperability, and enterprise deployment flexibility](https://daily.dev/posts/clickhouse-expands-collaboration-with-microsoft-bringing-fabric-integration-deeper-onelake-interop-bifbf9a3m) · ClickHouse · 0 upvotes · 0 comments
- [Microsoft Fabric IQ adds ‘semantic intelligence’ layer to Fabric](https://daily.dev/posts/microsoft-fabric-iq-adds-semantic-intelligence-layer-to-fabric-th0ejlsnd) · InfoWorld · 0 upvotes · 0 comments

---

Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#mcp](https://daily.dev/tags/mcp), [#clickhouse](https://daily.dev/tags/clickhouse), [#power-bi](https://daily.dev/tags/power-bi), [#microsoft-fabric](https://daily.dev/tags/microsoft-fabric)

[View this post on daily.dev](https://daily.dev/posts/microsoft-fabric-becomes-a-context-layer-for-ai-agents-with-clickhouse-added-for-fast-analytics-of5kmyizq)

```json
{"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]}
{"@context":"https://schema.org","@type":"TechArticle","headline":"Microsoft Fabric becomes a context layer for AI agents, with ClickHouse added for fast analytics","url":"https://daily.dev/posts/microsoft-fabric-becomes-a-context-layer-for-ai-agents-with-clickhouse-added-for-fast-analytics-of5kmyizq","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/microsoft-fabric-becomes-a-context-layer-for-ai-agents-with-clickhouse-added-for-fast-analytics-of5kmyizq"},"datePublished":"2026-09-30T14:52:11.680Z","dateModified":"2026-09-30T14:52:51.719Z","description":"Microsoft used FabCon and SQLCon in Barcelona to reposition Fabric as a context layer for enterprise AI agents. Fabric IQ, its context layer, now feeds...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/9de75ec2d271e8f2637e71934814ac24?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/9de75ec2d271e8f2637e71934814ac24?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Collections","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Collections","logo":"https://media.daily.dev/image/upload/s--fk_6ycEi--/f_auto,q_auto/v1780996001/logos/collections?_a=BAMAMiWQ0","url":"https://daily.dev/sources/collections"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/microsoft-fabric-becomes-a-context-layer-for-ai-agents-with-clickhouse-added-for-fast-analytics-of5kmyizq","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":2},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,mcp,clickhouse,power-bi,microsoft-fabric","timeRequired":"PT3M"}
{"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Collections","item":"https://daily.dev/sources/collections"},{"@type":"ListItem","position":3,"name":"Microsoft Fabric becomes a context layer for AI agents, with ClickHouse added for fast analytics"}]}
{"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/microsoft-fabric-becomes-a-context-layer-for-ai-agents-with-clickhouse-added-for-fast-analytics-of5kmyizq#faq","mainEntity":[{"@type":"Question","name":"What is Fabric IQ and how do AI agents access it?","acceptedAnswer":{"@type":"Answer","text":"Fabric IQ is Microsoft Fabric's context layer that feeds Microsoft 365 Copilot by default and exposes data to outside agents via MCP. Fabric IQ MCP is generally available with read-only tools, and ontologies layering business rules onto semantic models moved further into preview, letting agents understand terms like 'customer' or 'active order' correctly. Teams grounding AI agents in real business context can follow Fabric IQ's rollout on daily.dev."}},{"@type":"Question","name":"How does the new ClickHouse integration in Microsoft Fabric work?","acceptedAnswer":{"@type":"Answer","text":"A native ClickHouse workload for Fabric is in public preview, provisioning a dedicated ClickHouse Cloud service that syncs OneLake tables as accelerated copies without custom ETL. It includes an embedded SQL console, governed read-only access for AI agents (Copilot Studio, GitHub Copilot, MCP tools), Power BI, and notebooks, and runs on dedicated Azure compute separate from Fabric capacity so agent queries don't slow dashboards. Anyone speeding up analytics on OneLake can track ClickHouse's preview features on daily.dev."}},{"@type":"Question","name":"Which coding tools can drive Microsoft's new Osmos-based data engineering agent?","acceptedAnswer":{"@type":"Answer","text":"The Osmos-based data engineering agent added to Fabric can be driven from GitHub Copilot, VS Code, Codex, and Claude Code, giving developers multiple entry points to automate data engineering tasks directly from familiar coding environments rather than a separate Fabric-only interface. Developers wiring coding assistants into data workflows can follow these integrations on daily.dev."}}]}
```

