dbt Semantic Layer MCP Server

A Model-Connector-Presenter (MCP) server for seamlessly querying the dbt Semantic Layer through Claude Desktop and other compatible AI assistants.

What is the dbt Semantic Layer?

The dbt Semantic Layer is a powerful feature that allows you to define metrics once in your dbt project and reuse them consistently across your entire data stack. It provides:

• A single source of truth for business metrics
• Consistent metric definitions across all data tools
• Simplified access to complex metrics for all team members

About This Project

This MCP server acts as a bridge between AI assistants (like Claude) and the dbt Semantic Layer, enabling you to:

• Query metrics directly through natural language conversations
• Explore available metrics and their definitions
• Analyze data with dimensional breakdowns and filters
• Visualize results within your AI assistant interface

Features

• 🔍 Metric Discovery: Browse and search available metrics in your dbt Semantic Layer
• 📊 Query Creation: Generate and execute semantic queries through natural language
• 🧮 Data Analysis: Filter, group, and order metrics for deeper insights
• 📈 Result Visualization: Display query results in an easy-to-understand format

Prerequisites

• A dbt Cloud account with Semantic Layer enabled
• API access to your dbt Cloud instance
• Node.js (v14 or later)

Installation

Via Smithery (Recommended)

The easiest way to install is via Smithery:

npx -y @smithery/cli install @TommyBez/dbt-semantic-layer-mcp --client claude

Usage

Once installed and configured, you can interact with the dbt Semantic Layer directly from Claude Desktop:

1. Ask about available metrics: “What metrics are available in my dbt Semantic Layer?”
2. Query specific metrics: “Show me monthly revenue for the last quarter grouped by product category”
3. Analyze trends: “What’s the week-over-week growth in user signups?”

Troubleshooting

If you encounter issues:

• Verify your API credentials are correct
• Ensure your dbt Cloud project has Semantic Layer enabled
• Check that your metrics are properly defined in your dbt project

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

• dbt Labs for creating the dbt Semantic Layer
• Smithery for the MCP deployment platform
• LiteMCP for the MCP development package