mcp-server-bigqueryA Model Context Protocol server that provides access to BigQuery

— (0)

Reviewed by Daniel Nikulshyn·Updated July 2026

Overview

A Model Context Protocol (MCP) server that provides access to BigQuery, enabling Large Language Models (LLMs) to inspect database schemas and execute queries. The server implements tools such as 'execute-query' for running SQL queries, 'list-tables' for listing tables, and 'describe-table' for schema description. It can be configured via command line arguments or environment variables, requiring a GCP project ID and location, with optional settings for dataset filtering, service account key file, and query timeout. The server can be installed via Smithery for use with Claude Desktop.

Key features

execute-query
list-tables
describe-table
command line argument configuration
environment variable configuration

Pricing

Model: Free
Category: MCP Servers
Rating: No reviews yet

Use cases

LLM Database Interaction

Use the MCP server to enable LLMs to execute SQL queries and inspect database schemas in BigQuery

Automated Data Analysis

Configure the server to provide LLMs with access to specific BigQuery datasets for automated data analysis

Pros & Cons

Pros

Enables LLMs to interact with BigQuery
Supports multiple configuration options
Implements multiple tools for database interaction

Cons

Requires GCP project ID and location
Optional configuration may be needed for datasets and authentication

Reviews

No reviews yet. Be the first!

Q&A

How do I install the server for use with Claude Desktop?

Install via Smithery with the command "npx @smithery/cli install mcp-server-bigquery-client-claude" and configure the mcpServers entry in Claude’s config file, pointing to the command "uvx mcp server bigquery" with the required env vars.

Asked by Yara Mansour · Feb 15, 2026

What tools does the server expose for LLM interaction?

It provides three tools: 'execute-query' to run SQL, 'list-tables' to list all tables, and 'describe-table' to get a table’s schema. These enable LLMs to inspect schemas and execute queries directly.

Asked by Diego Fernández · Jan 24, 2026

Can I limit the server to specific BigQuery datasets?

Yes, use the BIGQUERY_DATASETS variable or repeat the --dataset argument to include only selected datasets. If omitted, the server will consider all datasets in the project.

Asked by Margaret Whitfield · Jan 19, 2026

How do I configure the mcp-server-bigquery for my GCP project?

Set the GCP project ID and location via command line arguments or environment variables (BIGQUERY_PROJECT and BIGQUERY_LOCATION). Optionally specify datasets (BIGQUERY_DATASETS), a service account key file (BIGQUERY_KEY_FILE), and a query timeout (BIGQUERY_TIMEOUT).

Asked by Vera Nováková · Dec 23, 2025