Google's open source MCP server that allows AI agents to interact with SQL databases.

The difference betweenm it and other MCP to SQL servers is that it takes out complexities such as connection pooling, authentication and observability so that the developer can solely focus on implementing the logic of his Agents.

The thing is that while it might be easy to code a MCP server, handling the underlying infrastructure is not. Instead, Google's toolbox requires just a few lines of Python code to sort that out.

The Toolbox acts as a middle man sitting between your agent and the database so that the agent can use tooling that accesses the database by not having direct unfiltered access to it. The following diagram will me the idea more clear:

That diagram roughly explains that the Toolbox facilitates communication between your IDE's coding agents or you AI orchestration layer such as LangChain and tools that can access the database. The gist here is that you don't have to write code for each AI orchestration layer but instead get access to the database through a unified api. That aside, the key advantages that the Toolbox offers are :

• Query in Plain English: Interact with your data using natural language right from your IDE. Ask complex questions like, "How many orders were delivered in 2024, and what items were in them?" without writing any SQL.
• Automate Database Management: Simply describe your data needs, and let the AI assistant manage your database for you. It can handle generating queries, creating tables, adding indexes, and more.
• Generate Context-Aware Code: Empower your AI assistant to generate application code and tests with a deep understanding of your real-time database schema.
• Slash Development Overhead: Radically reduce the time spent on manual setup and boilerplate. MCP Toolbox helps streamline lengthy database configurations, repetitive code, and error-prone schema migrations.

The server is available in the following formats:

• Binary
• Container image
• Homebrew
• Compiled from source
• Gemini CLI Extension

To write code you and integrate it with your LangChain / LangGraph application, you first have to install the Toolbox LangChain SDK:

pip install toolbox-langchain

and then load the Client:

from toolbox_langchain import ToolboxClient

# update the url to point to your server
async with ToolboxClient("http://127.0.0.1:5000") as client:

# these tools can be passed to your application!
tools = client.load_toolset()

While the Toolbox is written in Go, it is accessible from Python and Javascript as well.

If on the other hand you want to use it from your IDE and not your code, there's integration guides on the official page for Postgresql, CouchDB, Spanner ec.

For instance to connect your IDE/Claude desktop to Neo4j using Toolbox in order to expose your developer assistant tools to a Neo4j instance, do:

Install Claude Code.

1. Create a .mcp.json file in your project root if it doesn’t exist.

2. Add the following configuration, replace the environment variables with your values, and save:

{
"mcpServers": {
"neo4j": {
    "command": "./PATH/TO/toolbox",
"args": ["--prebuilt","neo4j","--stdio"],
"env": {
"NEO4J_URI": "",
"NEO4J_DATABASE": "",
"NEO4J_USERNAME": "",
"NEO4J_PASSWORD": ""
}

3. Restart Claude code to apply the new configuration.

Your AI tool is now connected to Neo4j using MCP. You can now try asking your AI assistant to get the graph schema or execute Cypher statements thanks to the Neo4j tools now available to the LLM:

• get_schema: extracts the complete database schema, including details about node labels, relationships, properties, constraints, and indexes.
• execute_cypher: executes any arbitrary Cypher statement.

Magic. At the end of the day as in any newly founded technology such as MCP, the first tooling approaches are crude but over time become more user friendly enabling easier interaction with whatever lies at the backend. Toolbox is one of those.

More Information

MCP Toolbox for Databases

Related Articles

LocalCode - A Perl-Based AI Coding Agent

To be informed about new articles on I Programmer, sign up for our weekly newsletter, subscribe to the RSS feed and follow us on Twitter, Facebook or Linkedin.

Comments

Make a Comment or View Existing Comments Using Disqus

or email your comment to: comments@i-programmer.info