Dbt Cli MCP

🚀 DBT CLI MCP Server

The DBT CLI MCP Server is a tool based on the Model Context Protocol (MCP) to enhance the functionality of dbt (Data Build Tool), a popular command-line tool for data building and testing.

🚀 Quick Start

The DBT CLI MCP Server is designed to enhance the capabilities of dbt. It supports key dbt operations, manages configuration files, and sets up environment variables automatically.

✨ Features

• Supported dbt Operations: It supports major dbt operations such as run, test, and compile.
• Configuration File Management: It can handle dbt_project.yml and profiles.yml files to ensure correct project configuration.
• Environment Variable Setup: It automatically configures the DBT_PROFILES_DIR environment variable to specify the location of the profiles file.

📦 Installation

Dependencies

• Git: Used for cloning the code repository and managing sub - modules.
• Python 3.8 or higher: Ensures compatibility.
• pip: Used for installing Python packages.

Clone the Repository

git clone https://github.com/your-repository.git
cd your-repository

Install Dependencies

pip install -r requirements.txt

💻 Usage Examples

Command - Line Interface (CLI)

Example Commands

• Run a dbt Project:

  python src/dbt_mcp_server/cli.py run --project-dir /path/to/project
  

• Test a dbt Project:

  python src/dbt_mcp_server/cli.py test --project-dir /path/to/project
  

Parameter Options

• --project-dir: Specifies the root directory of the dbt project, which must be an absolute path.
• --profile: Specifies the profile name in the dbt configuration file to be used.

Environment Variables

Example Environment Variable Setup

export DBT_PROFILES_DIR="/path/to/dbt_project"

📚 Documentation

Integration with MCP Client

Example Configuration File

Define the configuration in profiles.yml:

jaffle_shop:
  target: dev
  outputs:
    dev:
      type: duckdb
      path: 'jaffle_shop.duckdb'
      threads: 24

Notes

⚠️ Important Note

• The project - dir must be an absolute path that contains dbt_project.yml and profiles.yml.
• Ensure that the profile name specified in dbt_project.yml matches the one in profiles.yml. If you encounter a "Configuration file not found" error, check the path.

Development Information

Integration Tests

Run all integration tests:

python integration_tests/run_all.py

Run specific tests:

python integration_tests/test_dbt_run.py

Sub - module Management

Update the sub - module repository:

git submodule update --remote dbt_integration_tests/jaffle_shop_duckdb

Initialize sub - modules:

git submodule update --init

📄 License

This project is licensed under the MIT License.