Quick Start¶

Create a new project¶

fab-bundle init --template medallion --name my-analytics
cd my-analytics

This creates a project with: - 3 lakehouses (bronze, silver, gold) - 3 ETL notebooks - 1 data pipeline with scheduling - 1 Spark environment - 1 data agent - Dev/staging/prod targets

Configure your capacity¶

Find your Fabric capacity GUID:

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

Update fabric.yml with your capacity ID:

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

Validate¶

fab-bundle validate

Plan (dry-run)¶

fab-bundle plan --target dev

Deploy¶

fab-bundle deploy --target dev

Check status¶

fab-bundle status --target dev
fab-bundle drift --target dev

Quick Start with MCP (GitHub Copilot or Claude Code)¶

If you use GitHub Copilot or Claude Code, you can manage Fabric conversationally instead of typing CLI commands.

1. Install with MCP support¶

pip install fabric-automation-bundles[mcp]

2. Authenticate¶

az login

3. Add the MCP server¶

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

{
  "servers": {
    "fab-bundle": {
      "command": "fab-bundle-mcp"
    }
  }
}

Claude Code — add to .claude/settings.json in your project root (or ~/.claude/settings.json for global):

{
  "mcpServers": {
    "fab-bundle": {
      "command": "fab-bundle-mcp"
    }
  }
}

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

{
  "mcpServers": {
    "fab-bundle": {
      "command": "fab-bundle-mcp"
    }
  }
}

4. Add AI instructions to your project (optional but recommended)¶

Copy examples/.github/copilot-instructions.md or examples/CLAUDE.md to your project. These give your AI assistant context about your project structure and available tools.

GitHub Copilot: Copy examples/.github/copilot-instructions.md to .github/copilot-instructions.md
Claude Code: Copy examples/CLAUDE.md to CLAUDE.md in your project root

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"

Claude Code will use the 12 MCP tools (fab_validate, fab_plan, fab_deploy, fab_status, fab_drift, fab_run, etc.) to execute your requests against the live Fabric API.

6. Combine with Fabric VS Code Extension¶

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

Edit notebooks in VS Code with Claude Code helping
Run cells on remote Fabric Spark compute
Use fab-bundle MCP for infrastructure management
All without leaving VS Code

See Development Workflows for detailed patterns.