Databricks Genie

What is Databricks Genie MCP Server?

Databricks Genie MCP Server is a Model Context Protocol (MCP) server designed to connect Large Language Models (LLMs) with the Databricks Genie API. This connection enables LLMs to interact with Databricks conversational agents, run SQL queries, and ask natural language questions directly within the Databricks environment.

How to use Databricks Genie MCP Server?

To use Databricks Genie MCP Server, you need Python 3.7+ and a Databricks workspace with the Genie API enabled, a personal access token, and permissions to access Genie spaces and run queries. After cloning the repository, set up a virtual environment, install dependencies, and configure a .env file with your Databricks host and token. You will also need to manually add Genie space IDs and their titles in the get_genie_space_id() function in main.py as the API does not currently provide a public endpoint for this. The server can be tested using the MCP inspector or by building and running Docker. It can also be integrated with Claude Desktop by installing the MCP server and adding it as a resource in Claude.

Key Features of Databricks Genie MCP Server

• Genie Space Management: List and fetch metadata (title, description) of Genie spaces available in your Databricks workspace.
• Natural Language Interaction: Start new and follow-up Genie conversations using natural language questions.
• SQL Query Execution: Retrieve SQL queries and result tables in a structured format.
• MCP Tools: Provides tools like get_genie_space_id(), get_space_info(space_id: str), ask_genie(space_id: str, question: str), and follow_up(space_id: str, conversation_id: str, question: str).

Use Cases of Databricks Genie MCP Server

• Data Exploration: LLMs can explore Databricks data using natural language, making data more accessible to non-technical users.
• Automated Reporting: Generate reports by programmatically querying Databricks through natural language commands.
• Conversational Analytics: Build conversational agents that can interact with Databricks Genie to provide insights and answer data-related questions in real-time.
• Integration with LLMs: Seamlessly integrate Databricks functionalities into various LLM applications for enhanced data interaction.

FAQ about Databricks Genie MCP Server

• Q: What are the prerequisites for running the server?
  • A: Python 3.7+, Databricks workspace with Genie API enabled, a personal access token, and necessary permissions.
• Q: How do I get my Databricks credentials?
  • A: Your Databricks instance URL (without https://) is the host. For the token, go to your Databricks workspace, click your username -> User Settings -> Developer tab -> Manage under "Access tokens" to generate a new token.
• Q: Why do I need to manually add Genie Space IDs?
  • A: Currently, the Databricks Genie API does not provide a public endpoint to list all available space IDs and titles, so manual input is required.
• Q: What are common troubleshooting tips?
  • A: Ensure the host does not include https://, verify your personal access token is valid, check for timeouts, and ensure your query is valid for the selected space.
• Q: How can I test the server?
  • A: You can test it using the MCP inspector (npx @modelcontextprotocol/inspector python main.py) or by building and running Docker. You can also integrate it with Claude Desktop.
• Q: What are the security considerations?
  • A: Keep your .env file secure, use minimal scope tokens with expiration, and avoid exposing the server in public-facing environments unless authenticated.

Databricks Genie MCP Server

A Model Context Protocol (MCP) server that connects to the Databricks Genie API, allowing LLMs to ask natural language questions, run SQL queries, and interact with Databricks conversational agents.

✨ Features

• List Genie spaces available in your Databricks workspace (Currently Manual/Using Resource)
• Fetch metadata (title, description) of a specific Genie space
• Start new Genie conversations with natural language questions
• Ask follow-up questions in ongoing Genie conversations
• Retrieve SQL and result tables in structured format

🧱 Prerequisites

• Python 3.7+
• Databricks workspace with:
  • Personal access token
  • Genie API enabled
  • Permissions to access Genie spaces and run queries

⚙️ Setup

1. Clone this repository

2. Create and activate a virtual environment (recommended):

 python -m venv .venv
 source .venv/bin/activate

Install dependencies:

pip install -r requirements.txt

Create a .env file in the root directory with the following variables:

DATABRICKS_HOST=your-databricks-instance.cloud.databricks.com # Don't add https
DATABRICKS_TOKEN=your-personal-access-token

📌 Manually Adding Genie Space IDs

Note:
At this time, the Databricks Genie API does not provide a public endpoint to list all available space IDs and titles. (afaik) As a workaround, you need to manually add the Genie space IDs and their titles in the get_genie_space_id() function in main.py.

🧪 Test the Server

You can test the MCP server using the inspector (optional but recommended):

npx @modelcontextprotocol/inspector python main.py

OR

You can directly build and run docker to test the server

💬 Use with Claude Desktop

Download Claude Desktop

Install Your MCP Server: From your project directory, run:

mcp install main.py

Once Server Installed

1. Connect in Claude

2. Open Claude Desktop

3. Click Resources → Add Resource

4. Select your Genie MCP Server

5. Start chatting with your data using natural language! 🎯

🧾 Obtaining Databricks Credentials

Host Your Databricks instance URL (e.g., your-instance.cloud.databricks.com) — do not include https://

Token

1. Go to your Databricks workspace

2. Click your username (top right) → User Settings

3. Under the Developer tab, click Manage under "Access tokens"

4. Generate a new token and copy it

🚀 Running the Server

python main.py

This will start the Genie MCP server over the stdio transport for LLM interaction.

🧰 Available MCP Tools

The following MCP tools are available:

Tool Description

1. get_genie_space_id() List available Genie space IDs and titles
2. get_space_info(space_id: str) Retrieve title and description of a Genie space
3. ask_genie(space_id: str, question: str) Start a new Genie conversation and get results
4. follow_up(space_id: str, conversation_id: str, question: str) Continue an existing Genie conversation

🛠️ Troubleshooting

Common Issues

• Invalid host: Ensure the host does not include https://

• Token error: Make sure your personal access token is valid and has access to Genie

• Timeout: Check if the Genie space is accessible and not idle/expired

• No data returned: Ensure your query is valid for the selected space

🔐 Security Considerations

• Keep your .env file secure and never commit it to version control

• Use minimal scope tokens with expiration whenever possible

• Avoid exposing this server in public-facing environments unless authenticated

Claude Desktop Screenshots