dbhub and mcp-duckdb-memory-server
These are complements rather than competitors: bytebase/dbhub provides a general-purpose MCP server for multiple SQL databases with production features, while IzumiSy/mcp-duckdb-memory-server offers a lightweight, in-memory analytical database option that could be used alongside it for different query patterns or as a specialized backend within a broader data pipeline.
About dbhub
bytebase/dbhub
Zero-dependency, token-efficient database MCP server for Postgres, MySQL, SQL Server, MariaDB, SQLite.
Implements just two core MCP tools—`execute_sql` and `search_objects`—with optional custom parameterized operations defined in TOML config, minimizing token overhead for AI context windows. Supports stdio, HTTP, and SSE transports for integration with Claude Desktop, VS Code, Cursor, and other MCP clients, plus includes a built-in web workbench for direct database interaction. Enforces safety guardrails including read-only mode, row limits, and query timeouts while enabling secure multi-database connections through SSH tunneling and SSL/TLS.
About mcp-duckdb-memory-server
IzumiSy/mcp-duckdb-memory-server
MCP Memory Server with DuckDB backend
Implements a knowledge graph memory system storing entities, observations, and relations in DuckDB with SQL-backed queries, enabling scalable persistence beyond JSON file limitations. Combines DuckDB's analytical query engine with Fuse.js for fuzzy entity matching, supporting complex conditional searches and transaction integrity. Integrates with the Model Context Protocol (MCP) ecosystem for Claude Desktop and other MCP clients via stdio transport.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work