ScyllaDB Is Now Supported in MCP Toolbox for Databases

Connect your AI agents to ScyllaDB using the new ScyllaDB integration in MCP Toolbox for Databases

ScyllaDB is now a supported database source in MCP Toolbox for Databases (as of v1.5.0). This means that your agents and applications can query your ScyllaDB clusters using the same framework they use to connect to Postgres, MySQL, BigQuery, and dozens of other databases.

This post walks through what was added, how to set it up, and how to connect it to your agent or framework of choice.

What is MCP Toolbox for Databases?

MCP Toolbox for Databases is an open source Model Context Protocol (MCP) server that Google built to connect AI agents and applications directly to databases. It supports 40+ databases, including PostgreSQL, MySQL, MongoDB, Redis, Spanner, BigQuery, and now ScyllaDB.

It serves two purposes:

• As a ready-to-use MCP server: You can instantly plug your IDE assistant (Claude Code, Gemini CLI, Codex, and others) into your database. That lets your team ask questions in plain English, explore schemas, and generate data-aware code without writing boilerplate.
• As a custom tools framework: You can define safe, structured CQL tools for your production AI agents using a simple YAML config. Toolbox handles all the connection pooling, authentication, and observability out of the box.

Toolbox has client SDKs for Python, JavaScript/TypeScript, Go, and Java.

The ScyllaDB Integration

The integration adds two new resource types to MCP Toolbox:

• scylladb source: Connects Toolbox to a ScyllaDB cluster over the CQL protocol. It’s built on the ScyllaDB Go driver.
• scylladb-cql tool: Executes a pre-defined CQL statement against a ScyllaDB source, with support for parameterized queries and template parameters.

Both self-hosted ScyllaDB clusters and ScyllaDB Cloud are supported.

Installing and Running MCP Toolbox

Toolbox ships as a single binary. The quickest way to install it on Linux (AMD64):

curl -L https://github.com/googleapis/mcp-toolbox/releases/latest/download/toolbox_linux_amd64 -o toolbox
chmod +x toolbox

For other installation options, see the docs.

Once installed, start it with your configuration file:

./toolbox --config tools.yaml

Connecting MCP Toolbox to ScyllaDB

The scylladb source is defined in tools.yaml. Here’s a minimal self-hosted configuration:

kind: source
name: my-scylladb-source
type: scylladb
hosts:
  - 127.0.0.1
keyspace: my_keyspace
username: ${SCYLLA_USER}
password: ${SCYLLA_PASSWORD}

Connecting to ScyllaDB Cloud

For ScyllaDB Cloud, set localDC to the datacenter name shown on the Connect tab of your cluster. This enables DC-aware, token-aware load balancing:

kind: source
name: my-scylladb-cloud-source
type: scylladb
hosts:
  - node-0.your-cluster.us-east-1.cloud.scylladb.com
  - ...
keyspace: my_keyspace
username: ${SCYLLA_USER}
password: ${SCYLLA_PASSWORD}
localDC: AWS_US_EAST_1

Defining CQL Tools for Your AI Agent

Here is a simple lookup tool definition:

kind: tool
name: search_users_by_email
type: scylladb-cql
source: my-scylladb-source
statement: |
  SELECT user_id, email, first_name, last_name, created_at
  FROM users
  WHERE email = ?
parameters:
  - name: email
    type: string

The ? placeholder is replaced at runtime, preventing CQL injection. For dynamic table names, use templateParameters with Go template syntax.

Connecting to Your AI Client or Framework

MCP Clients (Claude Code, Gemini CLI, etc.)

{
  "mcpServers": {
    "toolbox": {
      "type": "http",
      "url": "http://127.0.0.1:5000/mcp"
    }
  }
}

Python Applications

from toolbox_core import ToolboxClient

async with ToolboxClient("http://127.0.0.1:5000") as client:
    tools = await client.load_toolset("user_tools")

To get started:

• ScyllaDB integration docs
• MCP Toolbox on GitHub
• ScyllaDB Cloud
• ScyllaDB University