Quick Start¶

This tutorial creates a Fabric project from a built-in template and deploys it to a development workspace. It assumes the tool is installed and authentication is configured; see Installation if not.

1. Create a new project¶

udp-cicd init --template medallion --name udp-analytics
cd udp-analytics

Project templates are rendered from Assets/templates/ using ${{ variable }} placeholder substitution. Five templates are available (see the Templates guide):

Template	Contents
`medallion`	A complete medallion-architecture project (see below).
`blank`	A minimal `udp.yml` with no resources, for starting from scratch.
`all-resource-types`	Reference catalogue declaring all 46 Fabric item types.
`realtime-governance`	IoT → Eventhouse → Eventstream → Activator + lakehouse, Data Agent, materialized lake views, and Azure governance; ships an Azure DevOps pipeline.
`streaming-lakehouse-governance`	Event Hub → Databricks → Lakehouse + Data Agent, with Azure governance; ships an Azure DevOps pipeline.

The medallion template creates a project containing:

Resource	Count	Purpose
Lakehouses	3	Bronze, silver, and gold layers
Notebooks	3	ETL processing per layer
Data pipelines	1	Orchestration with scheduling
Spark environments	1	Shared compute configuration
Data agents	1	Conversational access to gold data
Targets	3	Dev, staging, and prod environment definitions

2. Configure your capacity¶

Find the Fabric capacity GUID:

az rest --method get \
  --url "https://api.fabric.microsoft.com/v1/capacities" \
  --resource "https://api.fabric.microsoft.com"

Update udp.yml with the capacity ID:

workspace:
  capacity_id: "your-capacity-guid-here"

3. Validate, plan, and deploy¶

Run the three commands in sequence:

Command	Purpose
`udp-cicd validate`	Parses `udp.yml`, resolves variables, and checks the definition against the schema. No API calls are made.
`udp-cicd plan --target dev`	Dry run. Compares desired state against actual workspace state and prints the create, update, and delete actions that a deploy would perform.
`udp-cicd deploy --target dev`	Executes the plan against the Fabric API and records the result in deployment state.

udp-cicd validate
udp-cicd plan --target dev
udp-cicd deploy --target dev

4. Check status and drift¶

udp-cicd status --target dev
udp-cicd drift --target dev

status reports what is deployed according to the recorded state. drift compares recorded state against the live workspace and reports out-of-band changes.

5. Quick start with MCP (GitHub Copilot or Claude Code)¶

With GitHub Copilot or Claude Code, Fabric can be managed conversationally instead of through CLI commands.

5.1 Install with MCP support¶

dotnet tool install --global udp-cicd-mcp

5.2 Authenticate¶

az login

5.3 Add the MCP server¶

GitHub Copilot -- add to .vscode/mcp.json in the project root (or VS Code user settings.json under "mcp.servers" for global use):

{
  "servers": {
    "udp-cicd": {
      "command": "udp-cicd-mcp"
    }
  }
}

Claude Code -- add to .claude/settings.json in the project root (or ~/.claude/settings.json for global use):

{
  "mcpServers": {
    "udp-cicd": {
      "command": "udp-cicd-mcp"
    }
  }
}

Claude Desktop -- add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS):

{
  "mcpServers": {
    "udp-cicd": {
      "command": "udp-cicd-mcp"
    }
  }
}

5.4 Add AI instructions to your project (optional, recommended)¶

Copy the instruction file for your assistant into the project. These files give the AI assistant context about the project structure and available tools.

Assistant	Source	Destination
GitHub Copilot	`examples/.github/copilot-instructions.md`	`.github/copilot-instructions.md`
Claude Code	`examples/CLAUDE.md`	`CLAUDE.md` (project root)

5.5 Start talking¶

You: "Create a new Fabric project for sales analytics"
You: "What capacities do I have?"
You: "Deploy to dev"
You: "Run the ingest notebook"
You: "Check for drift in prod"
You: "Show me what's deployed"

The assistant uses the 14 MCP tools (udp_validate, udp_plan, udp_deploy, udp_status, udp_drift, udp_run, and others) to execute requests against the live Fabric API.

5.6 Combine with the Fabric VS Code extension¶

For the best experience, also install the Fabric Data Engineering VS Code Extension. This combination supports:

Editing notebooks in VS Code with AI assistance
Running cells on remote Fabric Spark compute
Managing infrastructure through the udp-cicd MCP server
Completing the full workflow without leaving VS Code

See Development Workflows for detailed patterns.