powerbi

Power BI Management & Automation

Power BI has no dedicated CLI. Use the Power BI REST API via az rest or Python. The user authenticates via az login.

Current context

• Azure account: !az account show --query "{name:name, id:id}" -o tsv 2>/dev/null || echo "not logged in — run: az login"

REST API via az rest

The Power BI REST API base URL is https://api.powerbi.com/v1.0/myorg.

For workspace-scoped operations, use: https://api.powerbi.com/v1.0/myorg/groups/{workspaceId}

Workspaces (Groups)

# List all workspaces the user has access to
az rest --method GET --url "https://api.powerbi.com/v1.0/myorg/groups"

# Get a specific workspace
az rest --method GET --url "https://api.powerbi.com/v1.0/myorg/groups/{workspaceId}"

# Create a workspace
az rest --method POST --url "https://api.powerbi.com/v1.0/myorg/groups?workspaceV2=true" --body '{"name": "New Workspace"}'

# Add a user to a workspace
az rest --method POST --url "https://api.powerbi.com/v1.0/myorg/groups/{workspaceId}/users" --body '{"emailAddress": "user@domain.com", "groupUserAccessRight": "Member"}'

Datasets (Semantic Models)

# List datasets in a workspace
az rest --method GET --url "https://api.powerbi.com/v1.0/myorg/groups/{workspaceId}/datasets"

# Get dataset details
az rest --method GET --url "https://api.powerbi.com/v1.0/myorg/groups/{workspaceId}/datasets/{datasetId}"

# Trigger dataset refresh
az rest --method POST --url "https://api.powerbi.com/v1.0/myorg/groups/{workspaceId}/datasets/{datasetId}/refreshes"

# Get refresh history
az rest --method GET --url "https://api.powerbi.com/v1.0/myorg/groups/{workspaceId}/datasets/{datasetId}/refreshes"

# Execute a DAX query against a dataset
az rest --method POST --url "https://api.powerbi.com/v1.0/myorg/groups/{workspaceId}/datasets/{datasetId}/executeQueries" --body '{
  "queries": [{"query": "EVALUATE SUMMARIZECOLUMNS('\''Date'\''[Year], \"Total Sales\", [Total Sales])"}],
  "serializerSettings": {"includeNulls": true}
}'

# Update dataset parameters
az rest --method POST --url "https://api.powerbi.com/v1.0/myorg/groups/{workspaceId}/datasets/{datasetId}/Default.UpdateParameters" --body '{
  "updateDetails": [{"name": "ServerName", "newValue": "newserver.database.windows.net"}]
}'

# Take over a dataset (change owner)
az rest --method POST --url "https://api.powerbi.com/v1.0/myorg/groups/{workspaceId}/datasets/{datasetId}/Default.TakeOver"

# Update data source credentials
az rest --method PATCH --url "https://api.powerbi.com/v1.0/myorg/gateways/{gatewayId}/datasources/{datasourceId}" --body '{
  "credentialDetails": {
    "credentialType": "OAuth2",
    "credentials": "{\"credentialData\":\"\"}",
    "encryptedConnection": "Encrypted",
    "encryptionAlgorithm": "None",
    "privacyLevel": "Organizational"
  }
}'

Reports

# List reports in a workspace
az rest --method GET --url "https://api.powerbi.com/v1.0/myorg/groups/{workspaceId}/reports"

# Get a specific report
az rest --method GET --url "https://api.powerbi.com/v1.0/myorg/groups/{workspaceId}/reports/{reportId}"

# Clone a report to another workspace
az rest --method POST --url "https://api.powerbi.com/v1.0/myorg/reports/{reportId}/Clone" --body '{"name": "Cloned Report", "targetWorkspaceId": "{targetWorkspaceId}"}'

# Rebind a report to a different dataset
az rest --method POST --url "https://api.powerbi.com/v1.0/myorg/groups/{workspaceId}/reports/{reportId}/Rebind" --body '{"datasetId": "{newDatasetId}"}'

# Export report (download .pbix)
az rest --method GET --url "https://api.powerbi.com/v1.0/myorg/groups/{workspaceId}/reports/{reportId}/Export" --output-file report.pbix

# Export report to file (PDF, PNG, PPTX)
az rest --method POST --url "https://api.powerbi.com/v1.0/myorg/groups/{workspaceId}/reports/{reportId}/ExportTo" --body '{
  "format": "PDF",
  "powerBIReportConfiguration": {
    "pages": [{"pageName": "ReportSection1"}]
  }
}'

Dashboards

# List dashboards in a workspace
az rest --method GET --url "https://api.powerbi.com/v1.0/myorg/groups/{workspaceId}/dashboards"

# List tiles on a dashboard
az rest --method GET --url "https://api.powerbi.com/v1.0/myorg/groups/{workspaceId}/dashboards/{dashboardId}/tiles"

Deployment Pipelines

# List deployment pipelines
az rest --method GET --url "https://api.powerbi.com/v1.0/myorg/pipelines"

# Get pipeline stages
az rest --method GET --url "https://api.powerbi.com/v1.0/myorg/pipelines/{pipelineId}/stages"

# Deploy content to next stage
az rest --method POST --url "https://api.powerbi.com/v1.0/myorg/pipelines/{pipelineId}/deployAll" --body '{
  "sourceStageOrder": 0,
  "options": {"allowOverwriteArtifact": true, "allowCreateArtifact": true}
}'

Admin operations

# List workspaces (admin scope — requires Power BI Admin role)
az rest --method GET --url "https://api.powerbi.com/v1.0/myorg/admin/groups?%24top=100"

# Get workspace info with datasets, reports, etc.
az rest --method POST --url "https://api.powerbi.com/v1.0/myorg/admin/workspaces/getInfo" --body '{"workspaces": ["{workspaceId}"]}'

# List capacities
az rest --method GET --url "https://api.powerbi.com/v1.0/myorg/capacities"

# Get refresh schedule
az rest --method GET --url "https://api.powerbi.com/v1.0/myorg/groups/{workspaceId}/datasets/{datasetId}/refreshSchedule"

For a full API reference, see api-reference.md.

Python automation

Using azure-identity + requests (recommended)

from azure.identity import DefaultAzureCredential
import requests

credential = DefaultAzureCredential()
token = credential.get_token("https://analysis.windows.net/powerbi/api/.default")

headers = {
    "Authorization": f"Bearer {token.token}",
    "Content-Type": "application/json"
}

# List workspaces
resp = requests.get("https://api.powerbi.com/v1.0/myorg/groups", headers=headers)
workspaces = resp.json()["value"]
for ws in workspaces:
    print(f"{ws['name']} ({ws['id']})")

Using semantic-link (in Fabric notebooks)

import sempy.fabric as fabric

# List datasets
datasets = fabric.list_datasets(workspace="<workspace-name>")

# Execute DAX query
result = fabric.evaluate_dax(
    dataset="<dataset-name>",
    dax_string='EVALUATE ROW("Test", 1 + 1)',
    workspace="<workspace-name>"
)

Using pbipy (community library)

pip install pbipy

from pbipy import PowerBI

pbi = PowerBI(bearer_token)

# List workspaces
workspaces = pbi.groups()

# Get datasets in a workspace
datasets = pbi.datasets(group=workspace)

# Trigger refresh
pbi.refresh_dataset(dataset, group=workspace)

Common automation scenarios

Bulk dataset refresh

# Refresh all datasets in a workspace
datasets = requests.get(f"{BASE}/groups/{ws_id}/datasets", headers=headers).json()["value"]
for ds in datasets:
    if ds.get("isRefreshable"):
        requests.post(f"{BASE}/groups/{ws_id}/datasets/{ds['id']}/refreshes", headers=headers)
        print(f"Triggered refresh: {ds['name']}")

Monitor refresh status

refreshes = requests.get(
    f"{BASE}/groups/{ws_id}/datasets/{ds_id}/refreshes?$top=1",
    headers=headers
).json()["value"]

latest = refreshes[0]
print(f"Status: {latest['status']}, Start: {latest.get('startTime')}, End: {latest.get('endTime')}")

Content promotion (dev -> test -> prod)

Use deployment pipelines for managed promotion:

# Deploy from Development (stage 0) to Test (stage 1)
az rest --method POST --url "https://api.powerbi.com/v1.0/myorg/pipelines/{pipelineId}/deployAll" --body '{"sourceStageOrder": 0}'

Theme development

Power BI themes are JSON files that control colors, typography, and visual formatting across an entire report. For full schema, starter templates, accessibility best practices, and application methods, see themes.md.

Quick reference:

# Apply theme via Power BI Embedded JS SDK
# report.applyTheme({ themeJson: themeObject });

# Apply theme via Semantic Link Labs (Fabric notebooks)
# rpt = labs.get_report("My Report", workspace="My Workspace")
# rpt.set_theme(theme_json=theme)

Theme files in PBIR projects live at: MyReport.Report/StaticResources/SharedResources/BaseThemes/.

PBIR & PBIP (Code-First Development)

PBIR (Enhanced Report Format) decomposes reports into individual JSON files per visual, page, and bookmark — enabling git-friendly, code-first Power BI development. PBIP (Power BI Project) is the project container pairing a PBIR report with a TMDL semantic model.

PBIR is the default format for new reports as of March 2026. For folder structure, JSON schemas, git workflow, and programmatic editing patterns, see pbir-reference.md.

Key capabilities:

• Edit report visuals, pages, and themes as individual JSON files
• Use VS Code with IntelliSense via Microsoft's published JSON schemas
• Git branching with per-file diffs (no monolithic blobs)
• Batch-modify visuals, copy pages between reports, CI/CD validation
• External tool compatibility: Tabular Editor, DAX Studio, ALM Toolkit

Frontend development (Embedding & Custom Visuals)

For building web applications with embedded Power BI content, creating custom visuals, or programmatic report authoring, see embedded.md.

Covers:

• Power BI Embedded (powerbi-client JS SDK) — embed reports/dashboards in web apps, handle events, runtime theme switching
• Report Authoring SDK (powerbi-report-authoring) — programmatic visual CRUD, data binding, layout manipulation
• Custom Visuals SDK (pbiviz) — create, develop, and package custom visuals
• Paginated Reports (RDL) — XML-based pixel-perfect reports
• Embed Token API — token generation and refresh patterns

Relationship with Fabric and DAX skills

• Fabric skill: Manages Fabric-specific items (lakehouses, warehouses, notebooks, Fabric pipelines). Power BI datasets/reports in Fabric are also accessible via this skill's REST API.
• DAX skill: Focuses on writing and optimizing DAX queries. Use executeQueries endpoint from this skill or evaluate_dax() from semantic-link to run DAX against Power BI datasets.
• PBIR crossover: PBIP projects contain both a .Report/ folder (PBIR, managed by this skill) and a .SemanticModel/ folder (TMDL, editable with the DAX skill or Tabular Editor).

Optional: Power BI MCP Servers

Power BI Modeling MCP (Microsoft)

Microsoft's powerbi-modeling-mcp (announced at Ignite 2025) enables AI agents to interact with semantic models:

• Read model structure, run DAX queries, create/modify measures, manage relationships
• Connects to Power BI Desktop, Fabric workspaces, and PBIP files
• Remote server available for cloud-hosted models

MCP Engine (Community)

• Repository: maxanatsko/pbi-desktop-mcp-public
• Connects to Power BI Desktop locally
• Read model structure, execute DAX, create/modify measures
• Compatible with Claude Desktop, Claude Code CLI, VS Code

Configuration example

{
  "mcpServers": {
    "powerbi": {
      "command": "npx",
      "args": ["-y", "@anthropic/powerbi-mcp"]
    }
  }
}

Note: Verify the exact package name and setup at the MCP server's documentation, as implementations are evolving.