MindsDB MCP server for AI agents

MindsDB is a powerful server that enables humans, AI, agents, and applications to get accurate answers across large-scale federated data sources. Its MCP (Model Context Protocol) server allows applications to connect, unify, and respond to questions over data spanning databases, data warehouses, and SaaS applications.

Installation Options

Using Docker Desktop

This is the fastest and recommended way to get started with MindsDB:

Make sure you have Docker Desktop installed on your system
Pull and run the MindsDB image:

docker pull mindsdb/mindsdb
docker run -p 47334:47334 -p 47335:47335 mindsdb/mindsdb

Using Docker

For more flexibility and customization:

docker pull mindsdb/mindsdb
docker run -p 47334:47334 -p 47335:47335 -v /path/to/storage:/root/mdb_storage mindsdb/mindsdb

You can customize the container by mounting storage volumes and setting environment variables.

Using PyPI

For installation via Python package manager:

pip install mindsdb
python -m mindsdb

Core Capabilities

Connecting Your Data

MindsDB allows you to connect to hundreds of enterprise data sources. These connections form the foundation for all other capabilities.

To connect a data source:

CREATE DATABASE my_database
WITH ENGINE = 'mysql',
PARAMETERS = {
    "user": "root",
    "password": "password",
    "host": "localhost",
    "port": 3306,
    "database": "example_db"
};

Unifying Your Data

Once connected, MindsDB enables querying data sources using SQL, as if they were all part of a single database.

Creating Views

Simplify data access with unified views across different sources:

CREATE VIEW combined_data AS
SELECT * FROM mysql_db.customers
JOIN postgres_db.orders ON customers.id = orders.customer_id;

Creating Knowledge Bases

Index and organize unstructured data for efficient retrieval:

CREATE KNOWLEDGE BASE product_docs
FROM files ('s3://mybucket/product_documentation/*.pdf');

Creating ML Models

Apply AI transformations to gain insights:

CREATE MODEL sales_forecaster
PREDICT sales
USING
    engine = 'openai',
    model_name = 'gpt-4',
    prompt_template = 'Forecast sales based on {{data}}';

Automating with Jobs

Schedule tasks for real-time processing:

CREATE JOB daily_sync
TASKS (
    UPDATE knowledge_base FROM source_data
)
SCHEDULE recurring every 24 hours;

Responding From Your Data

Configuring Agents

Set up specialized agents to answer questions:

CREATE AGENT sales_assistant
USING
    model = 'gpt-4',
    knowledge_base = 'sales_data',
    allowed_tools = ['query_database', 'calculate'];

Using the MCP Protocol

Connect to MindsDB through the Model Context Protocol for seamless interaction:

from mcp_client import MCPClient

# Connect to MindsDB MCP server
client = MCPClient("localhost:47335")

# Send a query to the agent
response = client.query(
    "What were our sales figures for Q1 2023?",
    agent="sales_assistant"
)
print(response)

Advanced Configuration

Server Settings

You can configure the MindsDB server by editing the config.json file or setting environment variables:

{
  "api": {
    "http": {
      "host": "0.0.0.0",
      "port": "47334"
    },
    "mysql": {
      "host": "0.0.0.0",
      "port": "47335",
      "user": "mindsdb",
      "password": "password"
    }
  },
  "storage_dir": "/path/to/storage"
}

Connecting Applications

To connect your applications to the MindsDB MCP server:

// JavaScript example
const { MCPClient } = require('mcp-client');

const client = new MCPClient({
  host: 'localhost',
  port: 47335,
  apiKey: 'your_api_key'
});

client.connect()
  .then(() => {
    return client.query('What insights can you provide from our customer data?');
  })
  .then(response => {
    console.log(response);
  });

Security Considerations

When deploying MindsDB in production:

Use strong passwords for all connections
Enable TLS/SSL for encrypted communications
Set up proper authentication mechanisms
Implement API keys for secure access
Regularly backup your MindsDB storage directory

How to install this MCP server

For Claude Code

To add this MCP server to Claude Code, run this command in your terminal:

claude mcp add-json "mindsdb" '{"command":"npx","args":["-y","mindsdb"]}'

See the official Claude Code MCP documentation for more details.

For Cursor

There are two ways to add an MCP server to Cursor. The most common way is to add the server globally in the ~/.cursor/mcp.json file so that it is available in all of your projects.

If you only need the server in a single project, you can add it to the project instead by creating or adding it to the .cursor/mcp.json file.

Adding an MCP server to Cursor globally

To add a global MCP server go to Cursor Settings > Tools & Integrations and click "New MCP Server".

When you click that button the ~/.cursor/mcp.json file will be opened and you can add your server like this:

{
    "mcpServers": {
        "mindsdb": {
            "command": "npx",
            "args": [
                "-y",
                "mindsdb"
            ]
        }
    }
}

Adding an MCP server to a project

To add an MCP server to a project you can create a new .cursor/mcp.json file or add it to the existing one. This will look exactly the same as the global MCP server example above.

How to use the MCP server

Once the server is installed, you might need to head back to Settings > MCP and click the refresh button.

The Cursor agent will then be able to see the available tools the added MCP server has available and will call them when it needs to.

You can also explicitly ask the agent to use the tool by mentioning the tool name and describing what the function does.

For Claude Desktop

To add this MCP server to Claude Desktop:

1. Find your configuration file:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
Linux: ~/.config/Claude/claude_desktop_config.json

2. Add this to your configuration file:

{
    "mcpServers": {
        "mindsdb": {
            "command": "npx",
            "args": [
                "-y",
                "mindsdb"
            ]
        }
    }
}

3. Restart Claude Desktop for the changes to take effect