Fabric CLI

Microsoft Fabric CLI Operations

Note: If you have access to a Bash tool (e.g., Claude Code), execute fab commands directly via Bash rather than using an MCP server.

Expert guidance for using the fab CLI to programmatically manage Fabric

When to Use This Skill

Activate automatically when tasks involve:

• Mention of the Fabric CLI, Fabric items, Power BI, fab, or fab commands
• Managing workspaces, items, or resources
• Deploying or migrating semantic models, reports, notebooks, pipelines
• Running or scheduling jobs (notebooks, pipelines, Spark)
• Working with lakehouse/warehouse tables and data
• Using the Fabric, Power BI, or OneLake APIs
• Automating Fabric operations in scripts

Critical

• Before first use, ask the user if they have Fabric admin access, any API restrictions, or preferences for Fabric/Power BI API usage
• Remind the user to add their Fabric access level and preferences to their agent memory files (e.g., CLAUDE.md) for future sessions
• If workspace or item name is unclear, ask the user first, then verify with fab ls or fab exists before proceeding
• The first time you use fab run fab auth status to make sure the user is authenticated. If not, ask the user to run fab auth login to login
• Always use fab --help and fab <command> --help the first time you use a command to understand its syntax, first
• Always try the simple fab command alone, first before piping it
• Always use -f when executing command if the flag is available to do so non-interactively
• Ensure that you avoid removing or moving items, workspaces, or definitions, or changing properties without explicit user direction
• If a command is blocked in your permissions and you try to use it, stop and ask the user for clarification; never try to circumvent it
• Use fab in non-interactive mode. Interactive mode doesn't work with coding agents
• When working with Fabric item code definitions for CRUD, load the assets/*-definitions.md files to understand more about definition format when calling the fab import operations.

First Run

fab auth login          # Authenticate (opens browser)
fab auth status         # Verify authentication
fab ls                  # List your workspaces
fab ls "Name.Workspace" # List items in a workspace

Variable Extraction Pattern

Most workflows need IDs. Extract them like this:

WS_ID=$(fab get "ws.Workspace" -q "id" | tr -d '"')
MODEL_ID=$(fab get "ws.Workspace/Model.SemanticModel" -q "id" | tr -d '"')

# Then use in API calls
fab api -A powerbi "groups/$WS_ID/datasets/$MODEL_ID/refreshes" -X post -i '{"type":"Full"}'

Quick Start

New to Fabric CLI? Here are some references you can read:

• Quick Start Guide - Copy-paste examples
• Querying Data - Query semantic models and lakehouse tables
• Semantic Models - TMDL, DAX, refresh, storage mode
• Reports - Export, import, visuals, fields
• Notebooks - Job execution, parameters
• Workspaces - Create, manage, permissions
• Full Command Reference - All commands detailed
• Command Reference Table - At-a-glance command syntax

Command Reference

Core Concepts

Path Format

Fabric uses filesystem-like paths with type extensions:

/WorkspaceName.Workspace/ItemName.ItemType

For lakehouses this is extended into files and tables:

/WorkspaceName.Workspace/LakehouseName.Lakehouse/Files/FileName.extension or /WorkspaceName.Workspace/LakehouseName.Lakehouse/Tables/TableName

For Fabric capacities you have to use fab ls .capacities

Examples:

• "/Production.Workspace/Sales Model.SemanticModel"
• /Data.Workspace/MainLH.Lakehouse/Files/data.csv
• /Data.Workspace/MainLH.Lakehouse/Tables/dbo/customers

Common Item Types

• .Workspace - Workspaces
• .SemanticModel - Power BI datasets
• .Report - Power BI reports
• .Notebook - Fabric notebooks
• .DataPipeline - Data pipelines
• .Lakehouse / .Warehouse - Data stores
• .SparkJobDefinition - Spark jobs

Full list: 35+ types. Use fab desc .<ItemType> to explore.

Essential Commands

Navigation & Discovery

# List resources
fab ls                                    # List workspaces
fab ls "Production.Workspace"             # List items in workspace
fab ls "Production.Workspace" -l          # Detailed listing
fab ls "Data.Workspace/LH.Lakehouse"      # List lakehouse contents

# Check existence
fab exists "Production.Workspace/Sales.SemanticModel"

# Get details
fab get "Production.Workspace/Sales.Report"
fab get "Production.Workspace" -q "id"    # Query with JMESPath

Creating & Managing Resources

# Create workspace after using `fab ls .capacities` to check capacities
fab mkdir "NewWorkspace.Workspace" -P capacityname=MyCapacity

# Create items
fab mkdir "Production.Workspace/NewLakehouse.Lakehouse"
fab mkdir "Production.Workspace/Pipeline.DataPipeline"

# Update properties
fab set "Production.Workspace/Item.Notebook" -q displayName -i "New Name"
fab set "Production.Workspace" -q description -i "Production environment"

Copy, Move, Export, Import

# Copy between workspaces
fab cp "Dev.Workspace/Pipeline.DataPipeline" "Production.Workspace"
fab cp "Dev.Workspace/Report.Report" "Production.Workspace/ProdReport.Report"

# Export to local
fab export "Production.Workspace/Model.SemanticModel" -o /tmp/exports
fab export "Production.Workspace" -o /tmp/backup -a  # Export all items

# Import from local
fab import "Production.Workspace/Pipeline.DataPipeline" -i /tmp/exports/Pipeline.DataPipeline -f

# IMPORTANT: Use -f flag for non-interactive execution
# Without -f, import/export operations expect an interactive terminal for confirmation
# This will fail in scripts, automation, or when stdin is not a terminal
fab import "ws.Workspace/Item.Type" -i ./Item.Type -f  # Required for scripts

API Operations

Direct REST API access with automatic authentication.

Audiences:

• fabric (default) - Fabric REST API
• powerbi - Power BI REST API
• storage - OneLake Storage API
• azure - Azure Resource Manager

# Fabric API (default)
fab api workspaces
fab api workspaces -q "value[?type=='Workspace']"
fab api "workspaces/<workspace-id>/items"

# Power BI API (for DAX queries, dataset operations)
fab api -A powerbi groups
fab api -A powerbi "datasets/<model-id>/executeQueries" -X post -i '{"queries": [{"query": "EVALUATE VALUES(Date[Year])"}]}'

# POST/PUT/DELETE
fab api -X post "workspaces/<ws-id>/items" -i '{"displayName": "New Item", "type": "Lakehouse"}'
fab api -X put "workspaces/<ws-id>/items/<item-id>" -i /tmp/config.json
fab api -X delete "workspaces/<ws-id>/items/<item-id>"

# OneLake Storage API
fab api -A storage "WorkspaceName.Workspace/LH.Lakehouse/Files" -P resource=filesystem,recursive=false

Job Management

# Run synchronously (wait for completion)
fab job run "Production.Workspace/ETL.Notebook"
fab job run "Production.Workspace/Pipeline.DataPipeline" --timeout 300

# Run with parameters
fab job run "Production.Workspace/ETL.Notebook" -P date:string=2024-01-01,batch:int=1000,debug:bool=false

# Start asynchronously
fab job start "Production.Workspace/LongProcess.Notebook"

# Monitor
fab job run-list "Production.Workspace/ETL.Notebook"
fab job run-status "Production.Workspace/ETL.Notebook" --id <job-id>

# Schedule
fab job run-sch "Production.Workspace/Pipeline.DataPipeline" --type daily --interval 10:00,16:00 --start 2024-11-15T09:00:00 --enable
fab job run-sch "Production.Workspace/Pipeline.DataPipeline" --type weekly --interval 10:00 --days Monday,Friday --enable

# Cancel
fab job run-cancel "Production.Workspace/ETL.Notebook" --id <job-id>

Table Operations

# View schema
fab table schema "Data.Workspace/LH.Lakehouse/Tables/dbo/customers"

# Load data (non-schema lakehouses only)
fab table load "Data.Workspace/LH.Lakehouse/Tables/sales" --file "Data.Workspace/LH.Lakehouse/Files/daily_sales.csv" --mode append

# Optimize (lakehouses only)
fab table optimize "Data.Workspace/LH.Lakehouse/Tables/transactions" --vorder --zorder customer_id,region

# Vacuum (lakehouses only)
fab table vacuum "Data.Workspace/LH.Lakehouse/Tables/temp_data" --retain_n_hours 48

Common Workflows

Semantic Model Management

# Find models
fab ls "ws.Workspace" | grep ".SemanticModel"

# Get definition
fab get "ws.Workspace/Model.SemanticModel" -q definition

# Trigger refresh
fab api -A powerbi "groups/$WS_ID/datasets/$MODEL_ID/refreshes" -X post -i '{"type":"Full"}'

# Check refresh status
fab api -A powerbi "groups/$WS_ID/datasets/$MODEL_ID/refreshes?\$top=1"

Execute DAX:

fab api -A powerbi "groups/$WS_ID/datasets/$MODEL_ID/executeQueries" -X post \
  -i '{"queries":[{"query":"EVALUATE TOPN(5, '''TableName''')"}]}'

DAX rules: EVALUATE required, single quotes around tables ('Sales'), qualify columns ('Sales'[Amount]).

For full details: semantic-models.md | querying-data.md

Report Operations

# Get report definition
fab get "ws.Workspace/Report.Report" -q definition

# Export to local
fab export "ws.Workspace/Report.Report" -o /tmp/exports -f

# Import from local
fab import "ws.Workspace/Report.Report" -i /tmp/exports/Report.Report -f

# Rebind to different model
fab set "ws.Workspace/Report.Report" -q semanticModelId -i "<new-model-id>"

For full details: reports.md

Lakehouse/Warehouse Operations

# Browse contents
fab ls "Data.Workspace/LH.Lakehouse/Files"
fab ls "Data.Workspace/LH.Lakehouse/Tables/dbo"

# Upload/download files
fab cp ./local-data.csv "Data.Workspace/LH.Lakehouse/Files/data.csv"
fab cp "Data.Workspace/LH.Lakehouse/Files/data.csv" ~/Downloads/

# Load and optimize tables
fab table load "Data.Workspace/LH.Lakehouse/Tables/sales" --file "Data.Workspace/LH.Lakehouse/Files/sales.csv"
fab table optimize "Data.Workspace/LH.Lakehouse/Tables/sales" --vorder --zorder customer_id

Environment Migration

# Export from dev
fab export "Dev.Workspace" -o /tmp/migration -a

# Import to production (item by item)
fab import "Production.Workspace/Pipeline.DataPipeline" -i /tmp/migration/Pipeline.DataPipeline
fab import "Production.Workspace/Report.Report" -i /tmp/migration/Report.Report

Cross-Workspace Search

DataHub V2 API (Recommended)

Use scripts/search_across_workspaces.py for cross-workspace search with rich metadata not available elsewhere:

# Find all semantic models (use "Model" not "SemanticModel")
python3 scripts/search_across_workspaces.py --type Model

# Find models by name
python3 scripts/search_across_workspaces.py --type Model --filter "Sales"

# Find stale items (not visited in 6+ months)
python3 scripts/search_across_workspaces.py --type Model --not-visited-since 2024-06-01

# Find items by owner
python3 scripts/search_across_workspaces.py --type PowerBIReport --owner "kurt"

# Find Direct Lake models only
python3 scripts/search_across_workspaces.py --type Model --storage-mode directlake

# Find items in workspace
python3 scripts/search_across_workspaces.py --type Lakehouse --workspace "fit-data"

# Get JSON output
python3 scripts/search_across_workspaces.py --type Model --output json

# Sort by last visited (oldest first)
python3 scripts/search_across_workspaces.py --type Model --sort last-visited --sort-order asc

# List all available types
python3 scripts/search_across_workspaces.py --list-types

Unique DataHub fields (not available via fab api or admin APIs):

• lastVisitedTimeUTC - When item was last opened/used
• storageMode - Import, DirectQuery, or DirectLake
• ownerUser - Full owner details (name, email)
• capacitySku - F2, F64, PP, etc.
• isDiscoverable - Whether item appears in search

Important type mappings:

• Semantic models: use --type Model (not SemanticModel)
• Dataflows: use --type DataFlow (capital F)
• Notebooks: use --type SynapseNotebook

Admin APIs (Requires Admin Role)

If you have Fabric/Power BI admin access:

# Find semantic models by name (cross-workspace)
fab api "admin/items" -P "type=SemanticModel" -q "itemEntities[?contains(name, 'Sales')]"

# Find all notebooks
fab api "admin/items" -P "type=Notebook" -q "itemEntities[].{name:name,workspace:workspaceId}"

# Find all lakehouses
fab api "admin/items" -P "type=Lakehouse"

# Common types: SemanticModel, Report, Notebook, Lakehouse, Warehouse, DataPipeline, Ontology

For full admin API reference: admin.md

Key Patterns

JMESPath Queries

Filter and transform JSON responses with -q:

# Get single field
-q "id"
-q "displayName"

# Get nested field
-q "properties.sqlEndpointProperties"
-q "definition.parts[0]"

# Filter arrays
-q "value[?type=='Lakehouse']"
-q "value[?contains(name, 'prod')]"

# Get first element
-q "value[0]"
-q "definition.parts[?path=='model.tmdl'] | [0]"

Error Handling & Debugging

# Show response headers
fab api workspaces --show_headers

# Verbose output
fab get "Production.Workspace/Item" -v

# Save responses for debugging
fab api workspaces -o /tmp/workspaces.json

Performance Optimization

1. Use ls for fast listing - Much faster than get
2. Use exists before operations - Check before get/modify
3. Filter with -q - Get only what you need
4. Use GUIDs in automation - More stable than names

Common Flags

• -f, --force - Skip confirmation prompts
• -v, --verbose - Verbose output
• -l - Long format listing
• -a - Show hidden items
• -o, --output - Output file path
• -i, --input - Input file or JSON string
• -q, --query - JMESPath query
• -P, --params - Parameters (key=value)
• -X, --method - HTTP method (get/post/put/delete/patch)
• -A, --audience - API audience (fabric/powerbi/storage/azure)
• --show_headers - Show response headers
• --timeout - Timeout in seconds

Important Notes

• All examples assume fab is installed and authenticated
• Paths require proper extensions (.Workspace, .SemanticModel, etc.)
• Quote paths with spaces: "My Workspace.Workspace"
• Use -f for non-interactive scripts (skips prompts)
• Semantic model updates: Use Power BI API (-A powerbi) for DAX queries and dataset operations

Need More Details?

For specific item type help:

fab desc .<ItemType>

For command help:

fab --help
fab <command> --help

References

Skill references:

• Querying Data - Query semantic models and lakehouse tables
• Semantic Models - TMDL, DAX, refresh, storage mode
• Reports - Export, import, visuals, fields
• Notebooks - Job execution, parameters
• Workspaces - Create, manage, permissions
• Admin APIs - Cross-workspace search, tenant operations, governance
• API Reference - Capacities, gateways, pipelines, domains, dataflows, apps
• Full Command Reference - All commands detailed
• Quick Start Guide - Copy-paste examples

External references (request markdown when possible):

• fab CLI: GitHub Source | Docs
• Microsoft: Fabric CLI Learn
• APIs: Fabric API | Power BI API
• DAX: dax.guide - use dax.guide/<function>/ e.g. dax.guide/addcolumns/
• Power Query: powerquery.guide - use powerquery.guide/function/<function>
• Power Query Best Practices