BigQuery MCP Server
Query BigQuery data and read dataset and table metadata through the Google-hosted BigQuery MCP server.
Publisher
Google Cloud
Product Details
BigQuery is Google Cloud’s autonomous data and AI platform, automating the data life cycle from data to AI to action. The BigQuery remote MCP server is a Google-hosted endpoint that lets AI applications run queries, get metadata, and list BigQuery resources. It is enabled when you enable the BigQuery API.
Users can list datasets and tables in a project, read dataset and table metadata, and run SQL queries. The read-only query tool accepts only SELECT statements, and execute_sql is the only tool that is not read-only. Results are capped at 3,000 rows and queries time out after three minutes by default. Sample requests include listing the datasets in a project, finding the top orders by volume in a dataset, and creating a forecast on a table.
Key Use Cases
Conversational Data Warehouse Exploration
Data analysts inspect project namespaces, retrieve table schemas, and run safe read-only SQL queries directly from an AI interface to speed up ad-hoc data discovery.
Automated Analytics Reporting
AI agents execute analytical queries against BigQuery data marts and synthesize high-volume metric outputs into concise executive business summaries.
Explore detailed deployment path
Requires Gemini. Access integration prerequisites, specialized agent configuration guides, and implementation documentation.