LucasHild/mcp-server-bigquery

Overview

Connects your AI assistant to BigQuery so you can ask questions about the data sitting in your warehouse, sales figures, marketing performance, product usage, and get an answer without writing SQL yourself. Built for businesses that already export data (from GA4, ad platforms, or internal systems) into BigQuery and want a faster way to query it. Requires a BigQuery project and service account already set up.

Best for

Best for
Businesses with data already flowing into BigQuery who want plain-English answers, not SQL

Use cases

• Ask for last month's sales or signups from your data warehouse
• Pull a marketing performance summary without writing a query
• Explore product usage data through natural-language questions
• Get a quick number for a report without opening a BI tool

How to use

Install

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

Tools exposed

• execute-query
• list-tables
• describe-table

Tested with

Claude Desktop, Claude Code, Cursor

Notes

An open-source Model Context Protocol server that exposes BigQuery as a tool for AI models. It enables models like Claude to query BigQuery databases by wrapping SQL operations in MCP endpoints. Built in Python and maintained by the community.

126 stars on GitHub. Last updated 2026-03-26. Licensed MIT.

Use cases

• Letting an AI assistant run BigQuery SQL queries on demand
• Integrating BigQuery data retrieval into MCP-compatible chat interfaces
• Building automated data exploration workflows that use natural language to query BigQuery

Pros

• Simple setup for developers already using BigQuery and MCP
• Lightweight Python implementation easy to extend or debug
• Directly bridges AI models with a major cloud data warehouse

Cons

• Requires manual BigQuery authentication and service account setup
• Limited to the MCP protocol, not a general-purpose BigQuery client
• Community-driven with modest adoption and documentation

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.