agentleFS
PatrickGallucciPatrickGallucci/fabric-skills/skills/fabric-iq/SKILL.md

fabric-iq

Skill17 starsChanged 9 months agoMITView on GitHub

Guide for working with Microsoft Fabric IQ (preview), the semantic intelligence workload for unified data and business vocabulary. Use when creating or managing ontology items, defining entity types, binding data to ontologies, creating relationship types, configuring data agents with ontology sources, working with Graph in Microsoft Fabric, managing Fabric IQ tenant settings, querying ontology graphs, generating ontologies from Power BI semantic models, or remediate Fabric IQ preview features. Covers ontology, graph, data agent, operations agent, and semantic model items.

  • Installs packages

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

10 rules. About 0 are common, 9 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Add entity types from the ribbon or canvas

    line 60
  2. Distinctive:

    Create relationship types between entity types and bind them to source data

    line 63
  3. Distinctive:

    Use the preview experience to explore entity instances and the ontology graph

    line 64
  4. Distinctive:

    Verify generated entity types, bindings, and relationship types

    line 73
  5. Distinctive:

    Create a Data agent item in your workspace

    line 80
  6. Distinctive:

    Add the ontology as a knowledge source

    line 81
  7. Very specific:

    Add agent instructions (e.g., Support group by in GQL)

    line 82
  8. Distinctive:

    Test queries in the agent chat to validate semantic grounding

    line 83
  9. Distinctive:

    Run the prereq validation script to check your environment:

    line 87
  10. Distinctive:

    Use the Fabric CLI for command-line operations:

    line 135

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Microsoft Fabric IQ

Fabric IQ (preview) is a Fabric workload for unifying data across OneLake and organizing it according to your business vocabulary. It exposes data to analytics, AI agents, and applications with consistent semantic meaning and context.

When to Use This Skill

  • Creating or managing ontology items in Fabric IQ
  • Defining entity types, properties, and relationship types
  • Binding data from lakehouses, eventhouses, or semantic models to ontologies
  • Generating ontologies from Power BI semantic models
  • Configuring Fabric data agents with ontology as a source
  • Working with Graph in Microsoft Fabric for traversals and graph queries
  • Enabling Fabric IQ tenant settings in the admin portal
  • Querying ontology graphs using the preview experience
  • Building operations agents that reason across business concepts
  • remediate ontology creation, data binding, or agent integration issues
  • Automating Fabric IQ items via REST API or PowerShell

Prerequisites

  1. A Fabric workspace with a Microsoft Fabric-enabled capacity (F2+ or P1+)
  2. Required tenant settings enabled (see tenant-settings.md)
  3. Data in OneLake (lakehouse tables), an eventhouse, or Power BI semantic models

Fabric IQ Items Overview

Fabric IQ contains five items that work together:

ItemPurposeShared With
Ontology (preview)Enterprise vocabulary and semantic layer — entity types, relationships, properties, data bindingsIQ only
Graph in Microsoft Fabric (preview)Native graph storage/compute for nodes, edges, traversals, path findingReal-Time Intelligence
Fabric data agent (preview)Conversational Q&A using generative AI, grounded in ontologyData Science
Operations agent (preview)AI agent to monitor real-time data and recommend actionsReal-Time Intelligence
Power BI semantic modelCurated analytics model for reporting and DAXPower BI

Choosing the Right Item

ScenarioUse
Cross-domain consistency, governance, AI agent groundingOntology
Relationship-heavy questions (impact chains, shortest paths)Graph
Trusted KPIs and fast visuals with dimensional modelingPower BI semantic model
Operational context, stateful twins, what-if simulationDigital twin builder (Real-Time Intelligence)

Step-by-Step Workflows

Workflow 1: Create an Ontology from OneLake

For the complete walkthrough with all field mappings, see ontology-workflows.md.

  1. Navigate to your Fabric workspace and select + New item > Ontology (preview)
  2. Name the ontology (letters, numbers, underscores only — no spaces or dashes)
  3. Add entity types from the ribbon or canvas
  4. Bind static or time series data from OneLake sources
  5. Set entity type keys (unique identifier properties)
  6. Create relationship types between entity types and bind them to source data
  7. Use the preview experience to explore entity instances and the ontology graph

Workflow 2: Generate an Ontology from a Semantic Model

For the complete walkthrough, see ontology-workflows.md.

  1. Navigate to your Power BI semantic model in Fabric
  2. Select Generate Ontology from the ribbon
  3. Choose workspace and name the ontology
  4. Verify generated entity types, bindings, and relationship types
  5. Configure any incomplete relationship bindings manually

Workflow 3: Connect an Ontology to a Data Agent

For the complete walkthrough, see ontology-workflows.md.

  1. Create a Data agent item in your workspace
  2. Add the ontology as a knowledge source
  3. Add agent instructions (e.g., Support group by in GQL)
  4. Test queries in the agent chat to validate semantic grounding

Workflow 4: Validate Tenant Prerequisites

Run the prereq validation script to check your environment:

./scripts/Validate-FabricIQPrereqs.ps1 -TenantId "your-tenant-id"

Key Concepts

Ontology Core Concepts

ConceptDescription
Entity typeRepresents a real-world concept (e.g., Customer, Truck, Sensor)
PropertyA fact about an entity type (e.g., name, email, temperature)
Entity type keyUnique identifier property for entity instances
Relationship typeSemantic connection between entity types (e.g., "drives", "has", "soldIn")
Data bindingConnects ontology definitions to concrete OneLake data sources
Ontology graphQueryable instance graph built from data bindings and relationships

Data Binding Types

TypeUse CaseExample
StaticDescriptive attributes that change infrequentlyStore locations, product catalog
Time seriesTimestamped observations in columnar formatSensor telemetry, temperature readings

Naming Constraints

ElementRules
Ontology nameLetters, numbers, underscores. No spaces or dashes
Entity type name1-26 chars, alphanumeric + hyphens + underscores, start/end alphanumeric
Property name1-26 chars, alphanumeric + hyphens + underscores, unique across entity types for same type

REST API Support

The Fabric REST API supports ontology CRUD operations:

OperationSupported
Create (without definition)Yes
Create (with payload/definition)Yes
Service principal supportYes
GetYes
UpdateYes
DeleteYes
ListYes

Use the Fabric CLI for command-line operations:

pip install ms-fabric-cli
fab auth login

remediate

For the full remediate guide, see remediate.md.

IssueQuick Fix
Unable to create ontology itemEnable all required tenant settings
Graph errors on new ontologyEnable User can create Graph (preview) tenant setting
Data agent 403 ForbiddenEnable Copilot and Azure OpenAI tenant settings
Generated ontology has no entity typesEnsure semantic model tables are visible (not hidden)
Generated ontology has no data bindingsCheck semantic model mode — Import mode not supported
Decimal properties return nullRecreate property as Double type
Aggregation queries fail in data agentAdd instruction: Support group by in GQL

References

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.