mcp-server-bigquery

Overview

The mcp-server-bigquery provides a bridge between large language models and Google BigQuery data warehouses. It allows LLM assistants to directly interact with structured data by executing SQL queries and retrieving results. The server implements three primary tools: execute-query for running SQL commands, list-tables for database discovery, and describe-table for schema inspection. This integration enables AI systems to analyze data patterns, generate reports, and answer questions based on actual BigQuery datasets without requiring manual data extraction or preprocessing.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• execute-query

  Executes a SQL query using BigQuery dialect

  Try:Show me the sales data for last monthTry:Calculate the average customer rating
• list-tables

  Lists all tables in the BigQuery database

  Try:What tables are available in this datasetTry:Show me all tables in the project
• describe-table

  Describes the schema of a specific table

  Try:Show me the schema of the customers tableTry:What columns are in the orders table

Comparable tools

Installation

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

{
  "mcpServers": {
    "bigquery": {
      "command": "uvx",
      "args": ["mcp-server-bigquery"],
      "env": {
        "BIGQUERY_PROJECT": "{{GCP_PROJECT_ID}}",
        "BIGQUERY_LOCATION": "{{GCP_LOCATION}}"
      }
    }
  }
}

claude mcp add bigquery --scope user --transport stdio -- uvx mcp-server-bigquery --project {PROJECT_ID} --location {LOCATION}

FAQ

What authentication methods does the server support?

The server supports service account key files via the --key-file argument or BIGQUERY_KEY_FILE environment variable. If not provided, it uses default GCP credentials.

Can I limit the server to specific datasets?

Yes, you can use the --dataset argument or BIGQUERY_DATASETS environment variable to specify which datasets to include. Multiple datasets can be specified by repeating the argument or using comma-separated values.

Compare mcp-server-bigquery with