mcp-server-duckdb and mcp-server-bigquery
These are ecosystem siblings—both are MCP server implementations for different SQL databases (BigQuery and DuckDB) that serve the same architectural role of providing database query capabilities to LLM applications, allowing users to choose based on their data infrastructure preference.
About mcp-server-duckdb
ktanaka101/mcp-server-duckdb
A Model Context Protocol (MCP) server implementation for DuckDB, providing database interaction capabilities
Exposes DuckDB through a unified `query` tool that accepts arbitrary SQL, enabling LLMs to perform analysis, schema inspection, and data operations on local databases. Supports read-only mode with DuckDB's native write protection and optional connection pooling for persistent state. Integrates with Claude Desktop and other MCP clients via stdio transport, with debugging support through the MCP Inspector.
About mcp-server-bigquery
LucasHild/mcp-server-bigquery
A Model Context Protocol server that provides access to BigQuery
Exposes three tools for LLM interaction: `execute-query` for BigQuery SQL execution, `list-tables` for schema discovery, and `describe-table` for inspecting column definitions. Communicates via stdio transport and integrates directly with Claude Desktop and Cursor through MCP configuration, supporting GCP authentication via service account keys or default credentials. Supports dataset filtering and configurable query timeouts to control execution scope and resource consumption.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work