BigQuery MCP Server

DatabasesTypeScript

Secure, read-only access to BigQuery datasets for LLMs

Available Tools

listDatasets

Lists all available datasets in the configured Google Cloud project

getDatasetSchema

Retrieves the schema of a specific dataset, showing tables and views

executeQuery

Executes a SQL query against BigQuery with read-only access

BigQuery MCP Server provides a secure bridge between Large Language Models and Google BigQuery databases. It enables natural language querying of your data, allowing you to ask questions in plain English and receive insights without writing SQL. The server maintains strict read-only access with processing limits to ensure data security while providing comprehensive access to tables and materialized views.

Overview

BigQuery MCP Server enables your AI assistants to directly query and analyze data stored in Google BigQuery. This integration allows for natural language conversations with your data, eliminating the need to manually write SQL queries or export data for analysis.

Features

Natural language querying of BigQuery datasets
Access to both tables and materialized views
Dataset schema exploration with clear resource type labeling
Secure read-only access with 1GB query processing limit by default
Simple authentication through Google Cloud

Prerequisites

Before setting up the BigQuery MCP Server, ensure you have:

Node.js 14 or higher installed
A Google Cloud project with BigQuery enabled
Either Google Cloud CLI installed or a service account key file
Claude Desktop (currently the only supported LLM interface)

Installation

Option 1: Quick Install via Smithery (Recommended)

The easiest way to install BigQuery MCP Server is through Smithery:

npx @smithery/cli install @ergut/mcp-bigquery-server --client claude

The installer will guide you through the configuration process, asking for your Google Cloud project ID and BigQuery location (defaults to us-central1). Once configured, Smithery will automatically update your Claude Desktop configuration.

Option 2: Manual Setup

If you prefer manual configuration:

Authenticate with Google Cloud using one of these methods:
- For development (using Google Cloud CLI):
```
gcloud auth application-default login
```
- For production (using a service account): Create and download a service account key file from the Google Cloud Console
Configure Claude Desktop by editing your claude_desktop_config.json file

Authentication and Permissions

The BigQuery MCP Server requires one of the following permission sets:

roles/bigquery.user (recommended)
OR both roles/bigquery.dataViewer and roles/bigquery.jobUser

Ensure your Google Cloud user account or service account has the appropriate permissions to access the datasets you want to query.

Usage

Once installed and configured, you can start using the BigQuery MCP Server by:

Opening Claude Desktop
Asking questions about your data in natural language

Example queries:

"What were our top 10 customers last month?"
"Show me the revenue trend for the past year by product category"
"What's the schema of the sales_data table?"

The server will translate your natural language questions into SQL queries, execute them against your BigQuery datasets, and return the results in a readable format.

Limitations

Currently only available in Claude Desktop (developer preview)
Connections limited to local MCP servers running on the same machine
Queries are read-only with a 1GB processing limit
Some complex view types might have limitations

Troubleshooting

If you encounter issues:

Verify your Google Cloud authentication is working correctly
Check that your service account has the necessary permissions
Ensure your project ID and location are correctly specified
Review Claude Desktop logs for any error messages

For persistent issues, visit the GitHub repository to report bugs or request features.