MCP Now
Keboola

Keboola

by keboola
GitHub

Build robust data workflows, integrations, and analytics on a single intuitive platform.

keboola
snowflake
server
configuration
desktop

Keboola MCP Server

CI codecov Keboola Explorer Server MCP server smithery badge

A Model Context Protocol (MCP) server for interacting with Keboola Connection. This server provides tools for listing and accessing data from Keboola Storage API.

Requirements

  • Keboola Storage API token
  • Snowflake Read Only Workspace

Installation

Installing via Smithery

To install Keboola Explorer for Claude Desktop automatically via Smithery:

1npx -y @smithery/cli install keboola-mcp-server --client claude

Manual Installation

First, clone the repository and create a virtual environment:

1git clone https://github.com/keboola/keboola-mcp-server.git
2cd keboola-mcp-server
3python3 -m venv .venv
4source .venv/bin/activate

Install the package in development mode:

1pip3 install -e .

For development dependencies:

1pip3 install -e ".[dev]"

Claude Desktop Setup

To use this server with Claude Desktop, follow these steps:

  1. Create or edit the Claude Desktop configuration file:

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

  2. Add the following configuration (adjust paths according to your setup):

1{
2  "mcpServers": {
3    "keboola": {
4      "command": "/path/to/keboola-mcp-server/.venv/bin/python",
5      "args": [
6        "-m",
7        "keboola_mcp_server",
8        "--api-url",
9        "https://connection.YOUR_REGION.keboola.com"
10      ],
11      "env": {
12        "KBC_STORAGE_TOKEN": "your-keboola-storage-token",
13        "PYTHONPATH": "/path/to/keboola-mcp-server/src",
14        "KBC_SNOWFLAKE_ACCOUNT": "your-snowflake-account",
15        "KBC_SNOWFLAKE_USER": "your-snowflake-user",
16        "KBC_SNOWFLAKE_PASSWORD": "your-snowflake-password",
17        "KBC_SNOWFLAKE_WAREHOUSE": "your-snowflake-warehouse",
18        "KBC_SNOWFLAKE_DATABASE": "your-snowflake-database",
19        "KBC_SNOWFLAKE_SCHEMA": "your-snowflake-schema",
20        "KBC_SNOWFLAKE_ROLE": "your-snowflake-role"
21      }
22    }
23  }
24}

Replace:

  • /path/to/keboola-mcp-server with your actual path to the cloned repository
  • your-keboola-storage-token with your Keboola Storage API token
  • YOUR_REGION with your Keboola region (e.g., north-europe.azure, etc.). You can remove it if your region is just connection explicitly
  • your-snowflake-account with your Snowflake account identifier
  • your-snowflake-user with your Snowflake username
  • your-snowflake-password with your Snowflake password
  • your-snowflake-warehouse with your Snowflake warehouse name
  • your-snowflake-database with your Snowflake database name
  • your-snowflake-schema with your Snowflake schema name
  • your-snowflake-role with your Snowflake role name

Note: If you are using a specific version of Python (e.g. 3.11 due to some package compatibility issues), you'll need to update the command into using that specific version, e.g. /path/to/keboola-mcp-server/.venv/bin/python3.11

Note: The Snowflake credentials can be obtained by creating a Read Only Snowflake Workspace in your Keboola project (the same project where you got your Storage Token). The workspace will provide all the necessary Snowflake connection parameters.

  1. After updating the configuration:
    • Completely quit Claude Desktop (don't just close the window)
    • Restart Claude Desktop
    • Look for the hammer icon in the bottom right corner, indicating the server is connected

Troubleshooting

If you encounter connection issues:

  1. Check the logs in Claude Desktop for any error messages
  2. Verify your Keboola Storage API token is correct
  3. Ensure all paths in the configuration are absolute paths
  4. Confirm the virtual environment is properly activated and all dependencies are installed
  5. Make sure the PYTHONPATH points to the src directory

Cursor AI Setup

To use this server with Cursor AI, you have two options for configuring the transport method: Server-Sent Events (SSE) or Standard I/O (stdio).

  1. Create or edit the Cursor AI configuration file:

    • Location: ~/.cursor/mcp.json

  2. Add one of the following configurations (or both) based on your preferred transport method:

Option 1: Using Server-Sent Events (SSE)

1{
2  "mcpServers": {
3    "keboola": {
4      "url": "http://localhost:8000/sse?storage_token=YOUR_STORAGE_TOKEN&snowflake_account=YOUR_ACCOUNT&snowflake_user=YOUR_USER&snowflake_password=YOUR_PASSWORD&snowflake_database=YOUR_DATABASE&snowflake_schema=YOUR_SCHEMA&snowflake_warehouse=YOUR_WAREHOUSE"
5    }
6  }
7}

Option 2: Using Standard I/O (stdio)

1{
2  "mcpServers": {
3    "keboola": {
4      "command": "/path/to/keboola-mcp-server/venv/bin/python",
5      "args": [
6        "-m",
7        "keboola_mcp_server",
8        "--transport",
9        "stdio",
10         "--api-url",
11         "https://connection.YOUR_REGION.keboola.com"
12      ],
13      "env": {
14        "KBC_STORAGE_TOKEN": "your-keboola-storage-token",
15        "KBC_SNOWFLAKE_ACCOUNT": "your-snowflake-account",
16        "KBC_SNOWFLAKE_USER": "your-snowflake-user",
17        "KBC_SNOWFLAKE_PASSWORD": "your-snowflake-password",
18        "KBC_SNOWFLAKE_DATABASE": "your-snowflake-database",
19        "KBC_SNOWFLAKE_SCHEMA": "your-snowflake-schema",
20        "KBC_SNOWFLAKE_WAREHOUSE": "your-snowflake-warehouse"
21      }
22    }
23  }
24}

Replace all placeholder values (your_*) with your actual Keboola and Snowflake credentials. These can be obtained from your Keboola project's Read Only Snowflake Workspace. Replace YOUR_REGION with your Keboola region (e.g., north-europe.azure, etc.). You can remove it if your region is just connection explicitly.

After updating the configuration:

  1. Restart Cursor AI
  2. If you use the sse transport make sure to start your MCP server. You can do so by running this in the activated virtual environment where you built the server: 
    /path/to/keboola-mcp-server/venv/bin/python -m keboola_mcp_server --transport sse --api-url https://connection.YOUR_REGION.keboola.com
  3. Cursor AI should be automatically detect your MCP server and enable it.

Available Tools

The server provides the following tools for interacting with Keboola Connection:

  • List buckets and tables
  • Get bucket and table information
  • Preview table data
  • Export table data to CSV
  • List components and configurations

Development

Run tests:

1pytest

Format code:

1black .
2isort .

Type checking:

1mypy .

License

MIT License - see LICENSE file for details.