mcp-bigquery-server and bigquery-mcp
About mcp-bigquery-server
ergut/mcp-bigquery-server
A Model Context Protocol (MCP) server that provides secure, read-only access to BigQuery datasets. Enables Large Language Models (LLMs) to safely query and analyze data through a standardized interface.
Implements MCP via Node.js with Google Cloud authentication (supporting both CLI and service account keys), exposing BigQuery tables and materialized views through natural language queries with built-in safety constraints (1GB processing limit, read-only access). Integrates exclusively with Claude Desktop via stdio transport, requiring configuration through the application's config file for direct LLM-to-database communication without manual SQL writing.
About bigquery-mcp
pvoo/bigquery-mcp
Practical MCP server for large BigQuery datasets. Supports vector search. Keep LLM context small while staying fast and allowing only safe read-only actions.
Implements MCP (Model Context Protocol) as a stdio-based server with dual-mode tools that minimize token usage by defaulting to lightweight responses—basic dataset listings return names only, detailed metadata loads on-demand. Enforces read-only safety through query validation (SELECT/WITH statements only) and automatic cost tracking with a ~$0.50 per-query billing cap. Integrates with BigQuery's Vertex AI embeddings for semantic vector search and supports dataset access control via allowlisting.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work