genai-toolbox

🥇Gold

The genai-toolbox is an open-source MCP server designed for integration with various databases, including BigQuery, MongoDB, and PostgreSQL. It helps developers to use AI capabilities across multiple database platforms, enhancing data management and analytics.

14,6381,1580Updated 1mo ago

Intermediate15 minutes to implementdatabase

Saves ~180 min per use

Quick InstallView Source

claude install googleapis/genai-toolbox

Works with:

ChatGPTGitHub Copilot

Overview

About This Skill

The genai-toolbox is an open-source MCP server maintained by Google APIs that enables AI agents to interact with multiple database platforms. It supports BigQuery, MongoDB, and PostgreSQL, allowing developers to leverage AI capabilities for database operations and analytics. The toolbox facilitates seamless integration between AI tools like Claude Code and Cursor with your existing database infrastructure. It provides a standardized interface for database queries and management across different database systems. This enables AI agents to perform data operations, analysis, and management tasks directly within supported database environments.

How to Use

["1. Install the genai-toolbox and ensure it is properly configured to connect to your database. Follow the installation instructions in the official documentation.","2. Use the prompt template to specify the database type, field name, value, and the fields you want to include in the output. For example: 'Using the genai-toolbox, connect to my MongoDB database and retrieve all documents where address.zip matches 11111. Format the output as a JSON array with the following fields: firstname, lastname, email.'","3. Run the prompt in your AI tool (e.g., Claude, ChatGPT) to execute the query. The AI will use the genai-toolbox to connect to the database and retrieve the requested data.","4. Review the output and ensure it meets your requirements. If necessary, refine your prompt and run the query again.","5. For better results, ensure your database is properly indexed for the fields you are querying. This can significantly improve query performance."]

Use Cases

Integrating AI with BigQuery to perform advanced data analytics and generate real-time insights.

Automating routine database queries using AI agents to reduce manual effort and increase efficiency.

Building custom AI solutions for data management that adapt to specific business requirements.

Enhancing data retrieval processes across MongoDB and PostgreSQL with intelligent query optimization.

Best For

GrowthRevOpsDevelopment

Setup & Installation

Quick Install

Terminal

claude install googleapis/genai-toolbox

Alternative Install (Git Clone)

git clone https://github.com/googleapis/genai-toolbox

Requirements

Claude Code or compatible AI agent
Works with: ChatGPT, GitHub Copilot

Quick Start Guide

Install the Skill

Copy the install command above and run it in your terminal.

Open Your AI Agent

Launch Claude Code, Cursor, or your preferred AI coding agent.

Try It Out

Use the prompt template or examples below to test the skill.

Customize

Adapt the skill to your specific use case and workflow.

Usage Examples

Prompt Template

Using the genai-toolbox, connect to my [DATABASE_TYPE] database and retrieve all documents where [FIELD_NAME] matches [VALUE]. Format the output as a JSON array with the following fields: [FIELD_LIST]. For example, if the database is MongoDB and the field is 'address.zip', the output should include 'firstname', 'lastname', and 'email'.

Example Output

Here is the output from a MongoDB database query using the genai-toolbox: 

```json
[
  {
    "firstname": "Bob",
    "lastname": "Smith",
    "email": "bob@smith.com",
    "address": {
      "street": "100 Main St",
      "city": "Anytown",
      "state": "MO",
      "zip": "11111"
    }
  },
  {
    "firstname": "Alice",
    "lastname": "Johnson",
    "email": "alice@johnson.com",
    "address": {
      "street": "200 Elm St",
      "city": "Othertown",
      "state": "NY",
      "zip": "11111"
    }
  }
]
```

This output shows two documents where the 'address.zip' field matches '11111'. The fields 'firstname', 'lastname', and 'email' are included as requested.