# Bricks Use

> A powerful Model Context Protocol (MCP) server for executing Databricks SQL queries and comparing table data.

- **Type:** MCP server
- **Install:** `agentstack add mcp-aymenfurter-bricks-use`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [aymenfurter](https://agentstack.voostack.com/s/aymenfurter)
- **Installs:** 0
- **Category:** [Databases](https://agentstack.voostack.com/c/databases)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [aymenfurter](https://github.com/aymenfurter)
- **Source:** https://github.com/aymenfurter/bricks-use

## Install

```sh
agentstack add mcp-aymenfurter-bricks-use
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

[](https://github.com/aymenfurter/bricks-use/actions/workflows/ci.yml)
[](https://www.python.org/downloads/)
[](https://opensource.org/licenses/MIT)
[](https://github.com/psf/black)
[](https://mypy.readthedocs.io/)
[](https://databricks.com/)

**A powerful Model Context Protocol (MCP) server for executing Databricks SQL queries and comparing table data.**

*⚠️ This project is purely meant for demo purposes - use at your own risk!*

---

## Table of Contents

- [Features](#features)
- [Quick Start](#quick-start)
- [Setup](#setup)
- [CLI Usage](#cli-usage)
- [MCP Tools](#mcp-tools)
- [VS Code Integration](#vs-code-integration)
- [License](#license)

## Features

| Feature | Description |
|---------|-------------|
| **Execute SQL Queries** | Run any SQL query on Databricks with configurable result limits |
| **Table Information** | Get detailed information about tables including schema and row counts |
| **Table Comparison** | Compare two tables by downloading their data and running CLI diff |
| **Quick Comparison** | Fast metadata-only comparison of tables |

## Quick Start

```bash
# 1. Clone and setup
git clone https://github.com/aymenfurter/bricks-use.git
cd bricks-use
python -m venv .venv && source .venv/bin/activate

# 2. Install dependencies
pip install -r requirements.txt

# 3. Configure environment (see setup section)
cp .env.example .env  # Edit with your credentials

# 4. Run the server or use CLI
python databricks_server.py  # For MCP server
# OR
./bricks query "SELECT * FROM my_table LIMIT 10"  # For CLI
```

## Setup

### Prerequisites

Python
3.11 or higher

Databricks
Workspace access

Token
Personal access token

### Environment Variables

Set the following environment variables or create a `.env` file:

```bash
# Databricks Configuration
DATABRICKS_SERVER_HOSTNAME=your-workspace.cloud.databricks.com
DATABRICKS_HTTP_PATH=/sql/1.0/warehouses/your-warehouse-id
DATABRICKS_ACCESS_TOKEN=your-personal-access-token

# Optional Settings
DATABRICKS_CATALOG=main                    # Defaults to 'main'
DATABRICKS_SCHEMA=default                  # Defaults to 'default'
DATABRICKS_TEMP_DIR=/tmp/databricks_mcp    # Temp directory
```

### Installation

1. **Create and activate a virtual environment:**
   ```bash
   python -m venv .venv
   source .venv/bin/activate  # On Windows: .venv\Scripts\activate
   ```

2. **Install dependencies:**
   ```bash
   pip install -r requirements.txt
   ```

## CLI Usage

Use the `./bricks` command-line tool for direct interaction:

```bash
# Execute SQL queries
./bricks query "SELECT * FROM my_table LIMIT 10"
./bricks query "SELECT COUNT(*) FROM users WHERE active = true" --limit 50

# Get table information
./bricks info my_table
./bricks info users --catalog production --schema analytics

# Compare tables
./bricks compare table1 table2
./bricks compare old_users new_users --quick
./bricks compare sales_2023 sales_2024 --catalog1 prod --schema1 sales

# Output options
./bricks query "SELECT * FROM table" --format json
./bricks info my_table --format table
```

## MCP Tools

| Tool | Purpose | Key Parameters |
|------|---------|----------------|
| `execute_query` | Execute SQL queries | `query`, `limit` |
| `get_table_info` | Get table metadata | `table_name`, `catalog`, `schema` |
| `compare_tables` | Full data comparison | `table1`, `table2`, `diff_lines` |
| `quick_compare_tables` | Metadata comparison | `table1`, `table2` |

---

### execute_query

Execute a SQL query on Databricks.

**Parameters:**
- `query` (str): SQL query to execute
- `limit` (int, optional): Maximum rows to return (default: 1000)

### get_table_info

Get information about a Databricks table.

**Parameters:**
- `table_name` (str): Name of the table
- `catalog` (str, optional): Catalog name
- `schema` (str, optional): Schema name

### compare_tables

Compare data between two tables by downloading full data and running diff.

**Parameters:**
- `table1` (str): First table name
- `table2` (str): Second table name
- `catalog1` (str, optional): Catalog for table1
- `schema1` (str, optional): Schema for table1
- `catalog2` (str, optional): Catalog for table2
- `schema2` (str, optional): Schema for table2
- `diff_lines` (int, optional): Number of diff context lines (default: 10)

### quick_compare_tables

Quick metadata-only comparison without downloading data.

**Parameters:**
- `table1` (str): First table name
- `table2` (str): Second table name
- `catalog1` (str, optional): Catalog for table1
- `schema1` (str, optional): Schema for table1
- `catalog2` (str, optional): Catalog for table2
- `schema2` (str, optional): Schema for table2

## VS Code MCP Integration

Add this configuration to your VS Code settings (`mcp.json`):

Click to expand VS Code configuration

```json
{
    "inputs": [
        {
            "type": "promptString",
            "id": "databricks_server_hostname",
            "description": "Databricks Server Hostname"
        },
        {
            "type": "promptString",
            "id": "databricks_http_path",
            "description": "Databricks HTTP Path"
        },
        {
            "type": "promptString",
            "id": "databricks_access_token",
            "description": "Databricks Access Token",
            "password": true
        },
        {
            "type": "promptString",
            "id": "databricks_catalog",
            "description": "Databricks Catalog (default: main)"
        },
        {
            "type": "promptString",
            "id": "databricks_schema",
            "description": "Databricks Schema (default: default)"
        }
    ],
    "servers": {
        "databricks": {
            "command": "python",
            "args": [
                "${workspaceFolder}/databricks_server.py"
            ],
            "env": {
                "PYTHONUNBUFFERED": "1",
                "DATABRICKS_SERVER_HOSTNAME": "${input:databricks_server_hostname}",
                "DATABRICKS_HTTP_PATH": "${input:databricks_http_path}",
                "DATABRICKS_ACCESS_TOKEN": "${input:databricks_access_token}",
                "DATABRICKS_CATALOG": "${input:databricks_catalog}",
                "DATABRICKS_SCHEMA": "${input:databricks_schema}"
            },
            "workingDirectory": "${workspaceFolder}"
        }
    }
}
```

---

## License

This project is licensed under the [MIT License](https://opensource.org/licenses/MIT).

**Made with ❤️ for the Databricks community**

## Source & license

This open-source MCP server is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [aymenfurter](https://github.com/aymenfurter)
- **Source:** [aymenfurter/bricks-use](https://github.com/aymenfurter/bricks-use)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** yes
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/mcp-aymenfurter-bricks-use
- Seller: https://agentstack.voostack.com/s/aymenfurter
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.