MCP-PostgreSQL-Ops and postgres-mcp
About MCP-PostgreSQL-Ops
call518/MCP-PostgreSQL-Ops
🔍Professional MCP server for PostgreSQL operations & monitoring: 30+ extension-independent tools for performance analysis, table bloat detection, autovacuum monitoring, schema introspection, and database management. Supports PostgreSQL 12-17.
Implements the Model Context Protocol (MCP) standard to integrate with AI assistants and language models, exposing PostgreSQL operations as callable tools through stdio transport. Leverages optional `pg_stat_statements` and `pg_stat_monitor` extensions for enhanced query analytics while maintaining core functionality on base PostgreSQL installations. Built with Python and containerized for Docker deployment, supporting RDS/Aurora environments with read-only permissions suitable for production use.
About postgres-mcp
neverinfamous/postgres-mcp
PostgreSQL MCP Server: Secure Administration & Observability Featuring Code Mode—One Tool Replacing All Specialized 232 Tools for up to 90% Token Savings. Includes Connection Pooling, HTTP/SSE, OAuth 2.1, Deterministic Error Handling and Full Support for 8 Extensions (citext, HypoPG, ltree, pgcrypto, pg_cron, pg_stat_kcache, pgvector & PostGIS).
Implements a JavaScript sandbox (Code Mode) that executes multi-step database operations locally within a single tool invocation, enabling AI agents to chain 232 PostgreSQL capabilities without round-trip communication overhead. Provides 20 observability resources for real-time schema, performance, and connection metrics, plus 19 guided prompts for query optimization and extension management. Targets AI assistants and agents via MCP with dual HTTP/SSE transport, schema introspection, migration tracking with SHA-256 deduplication, and granular OAuth 2.1 scopes (`read`, `write`, `admin`, `db:*`, `table:*:*`).
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work