🛠️ dbt MCP Server
About
The dbt MCP Server is a minimal, extensible MCP (Model Context Protocol) server that provides a unified JSON-based interface for running dbt models, tests, and other dbt operations. It can be triggered via CLI, HTTP API, or an AI interface like Claude, and includes Slack…
Details
- Author
- kannandreams
- GitHub stars
- 3
- Downloads
- 274
- Categories
- Search
Jump to
- Run dbt models using a clean JSON spec
- Accept CLI or HTTP API calls
- Slack alerts for model success or failure
- Compatible with AI tools like Claude
- Extensible and minimal design
- Unified interface for triggering dbt operations
Setting up with Highlight
This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
🛠️ dbt MCP ServerCommand (node, npx, python, etc.)Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
From the repository
Run the server locally with uvicorn mcp_dispatcher:app --reload --port 8000 or use Docker with docker build -t dbt-mcp-server . then docker run -p 8000:8000 dbt-mcp-server. Trigger dbt models by executing python mcp_dispatcher.py mcp_request.json with a JSON spec. Integrate with Claude via python claude_client_example.py. Set the SLACK_WEBHOOK_URL environment variable to receive Slack alerts.
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"\ud83d\udee0\ufe0f dbt mcp server": {
"dbt-mcp-server": {
"command": "docker",
"args": [
"build",
"-t",
"dbt-mcp-server",
"."
]
}
}
}
}McpServers
{
"dbt-mcp-server": {
"command": "docker",
"args": [
"build",
"-t",
"dbt-mcp-server",
"."
]
}
}🛠️ dbt MCP Server
This is a minimal, extensible MCP (Model Control Protocol) server to run dbt models with a unified interface.If you like this repo, leave a ⭐️ — it means a lot!
Overview
A unified JSON-based workflow for triggering dbt models, tests, and other dbt operations, triggered via CLI, API, or an AI interface like Claude. Includes Slack notifications for task completions.🚀 Getting Started
✅ Ready to plug into AI tools like Claude for prompt-driven automation!
Run Locally with UV
uvicorn mcp_dispatcher:app --reload --port 8000
Run with Docker
docker build -t dbt-mcp-server .
docker run -p 8000:8000 dbt-mcp-server
5. Trigger MCP from Claude (supported client)
python claude_client_example.py
📣 Slack Alerts
Set yourSLACK_WEBHOOK_URL to get alerts on model success or failure.
🔌 Usage
python mcp_dispatcher.py mcp_request.json
{
"mcpServers": {
"fetch": {
"command": "node",
"args": [
"{ABSOLUTE PATH TO FILE HERE}/dist/index.js"
]
}
}
}
Features
- Run dbt models using a clean JSON spec - Accept CLI or HTTP API calls - Slack alerts for model status - Compatible with Claude🛠️ Contributing
If you're interested in contributing, feel free to open an issue to start a discussion or directly submit a pull request. Whether it's code, ideas, or feedback—all contributions are welcome!License
This MCP server is licensed under the MIT License. This means you are free to use, modify, and distribute the software, subject to the terms and conditions of the MIT License. For more details, please see the LICENSE file in the project repository.Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
