pgmcp
Active·★ 541·NOASSERTION·Updated 2026-05-26
PGMCP connects AI assistants to any PostgreSQL database through natural language queries. It provides secure read-only access, automatic streaming, and supports multiple output formats. It works with various MCP-compatible clients like Cursor and Claude Desktop.
pgmcp sits in the Dev Tooling category, so it is usually better judged by workflow fit, deployment style, and team needs than by isolated features alone. Based on the available metadata, it leans most clearly toward Natural Language to SQL. A representative use case is Query database using natural language. The listed license is NOASSERTION. GitHub stars are currently 541. Relevant tags include agent, agentic-ai.
#agent#agentic-ai#ai#analytics#artificial-intelligence#data-analysis#database#kong
01
Features
01Natural Language to SQL
02Automatic Streaming
03Safe Read-Only Access
04Text Search
05Multiple Output Formats
02
Why choose it
+Natural Language to SQL
+Automatic Streaming
+Query database using natural language
+Supports 5 platforms or runtime environments.
03
Trade-offs
!The final decision should still be made through side-by-side comparison with similar tools.
04
Compatibility
Linux
Linux (Binary)
Verified via docs
macOS
macOS (Binary)
Verified via docs
Windows
Windows (Binary)
Verified via docs
Docker
Docker
Verified via docs
Kubernetes
Kubernetes
Verified via docs
05
Quick start
1
$ docker run -e DATABASE_URL='postgres://user:pass@host:5432/db' -p 8080:8080 ghcr.io/subnetmarco/pgmcp:latest
06
Use cases
↳Query database using natural language
↳Search across all text columns
↳Integrate with MCP clients like Cursor and Claude Desktop
07
Alternatives
Comments
Log in to leave a comment
No comments yet. Be the first!