name: datafactory.architecture description: "DataFactory.MCP project architecture, component layers, design patterns, and extension points. Use when working on any code in this repo."

DataFactory.MCP Architecture

For the full architecture reference, see docs/ARCHITECTURE.md.

Project Structure

DataFactory.MCP/             — Main MCP server (stdio transport, entry point)
DataFactory.MCP.Core/        — Core library (all business logic)
  ├── Abstractions/          — Interfaces and contracts
  ├── Configuration/         — App configuration classes
  ├── Extensions/            — Extension methods and DI registration
  ├── Infrastructure/        — Cross-cutting concerns (HTTP, logging)
  ├── Models/                — Data models and DTOs
  ├── Parsing/               — Input parsing utilities
  ├── Resources/             — Embedded resources
  ├── Services/              — Fabric API service clients
  ├── Tools/                 — MCP tool implementations
  └── Validation/            — Input validation
DataFactory.MCP.Http/        — HTTP transport layer (AspNetCore)
DataFactory.MCP.Tests/       — Test suite (xUnit)
DataFactory.WindowsMCP/      — Windows-specific MCP implementation
claude-skills/               — Claude skill definitions (RAG pattern)
docs/                        — Feature documentation
evals/                       — Evaluation test scenarios

Layered Architecture

AI Chat Interface (VS Code, Visual Studio)
        │ MCP Protocol (JSON-RPC over stdio)
        ▼
┌─ MCP Tools Layer ──────────────────────────────┐
│  AuthenticationTool, GatewayTool,               │
│  ConnectionsTool, WorkspacesTool,               │
│  Dataflow/*, Pipeline/*, CopyJob/*,             │
│  CapacityTool, CreateConnectionUITool           │
├─ Core Services Layer ──────────────────────────┤
│  AuthenticationService, FabricGatewayService,   │
│  FabricConnectionService, FabricWorkspaceService│
│  FabricDataflowService, FabricPipelineService,  │
│  FabricCopyJobService, FabricCapacityService,   │
│  ValidationService, DataTransformationService   │
├─ Infrastructure Layer ─────────────────────────┤
│  AuthenticatedHttpHandler, ApiVersionConfig,    │
│  McpSessionAccessor, BackgroundTasks            │
├─ Abstractions Layer ───────────────────────────┤
│  Interfaces for all services                    │
├─ Models Layer ─────────────────────────────────┤
│  DTOs, request/response types, enums            │
└─────────────────────────────────────────────────┘
        │ Fabric REST APIs (with Azure AD tokens)
        ▼
    Microsoft Fabric / Power BI Service

Key Design Principles

• Separation of Concerns: Tools handle MCP protocol; Services handle Fabric API calls
• Dependency Injection: All services registered via DI with proper lifetimes
• Async-First: All I/O operations use async/await
• Configuration-Driven: Feature flags control tool availability
• Extensibility: New tools/services can be added without modifying existing code

Feature Documentation

Domain Knowledge

For Data Factory / M language / Dataflow patterns, see claude-skills/:

• datafactory-core.md — M basics, MCP tools, connection discovery
• datafactory-destinations.md — Output destinations, DataDestination patterns
• datafactory-performance.md — Query tuning, chunking, query folding
• datafactory-advanced.md — Fast Copy, Action.Sequence, Modern Evaluator