Postgres Mcp
io.github.YawLabs/postgres-mcp is a MCP server that lets your agent query PostgreSQL, introspect schema, EXPLAIN plans, and run health checks.
About
The PostgreSQL MCP server (io.github.YawLabs/postgres-mcp) gives AI assistants controlled access to query, introspect schema, run EXPLAIN, and perform health checks on Postgres databases developers rely on for SaaS and API products. Use it during build backend work when migrating tables or tuning ORM queries, and again in operate monitoring when incidents need quick read-only diagnosis. It targets developers who already run Postgres locally or on a managed host and want Claude Code or Cursor to reason over real structure instead of guessing from migrations alone. Intermediate complexity applies because you must scope credentials, read-only vs write access, and safe SQL patterns. It is a database MCP integration, not a migration framework; pair it with your existing ORM and CI tests before destructive changes.
- Execute SQL queries against PostgreSQL from the agent
- Schema introspection for tables, columns, and relationships
- EXPLAIN support for query plan debugging
- Database health checks for operate-time triage
- @yawlabs/postgres-mcp version 0.6.6 with stdio transport
Postgres Mcp by the numbers
- Data as of Jul 7, 2026 (Skillselion catalog sync)
claude mcp add postgres-mcp -- npx -y @yawlabs/postgres-mcpAdd your badge
Show developers this MCP server is listed on Skillselion. Paste this into your README.
| Package | @yawlabs/postgres-mcp |
|---|---|
| Transport | STDIO |
| Auth | None |
What it does
Run SQL, inspect schema, EXPLAIN plans, and health-check Postgres from your agent while you build or debug production data issues.
Who is it for?
Best when you're shipping on Postgres and want agent-assisted query tuning and schema exploration with a configured connection string.
Skip if: Greenfield projects with no Postgres instance yet or teams that forbid any agent database access for compliance reasons.
What you get
You get live queries, schema context, explain output, and health signals inside the same session where you edit backend code.
- Agent-executed SQL and result sets for debugging
- Schema metadata the model can reference while coding
- EXPLAIN output and health check summaries for tuning or ops
By the numbers
- Package version 0.6.6 published as @yawlabs/postgres-mcp
- stdio MCP transport in server manifest
- Four described capabilities: query, schema introspection, explain, health checks
Recommended MCP Servers
A2dbSafe, read-only database access for AI agents.2
AdoMcpDatabase MCP server for schema discovery, comments, and SQL queries.1
Afgong Sqlite Mcp ServerExplore your Messages SQLite database to browse tables and inspect schemas with ease.
AI ERDAI-powered ERD design tool.
AirtableCreate, read, update, and delete records in Airtable bases and tables29
AirtableRead and write access to Airtable database schemas, tables, and records.455How it compares
Postgres-connected MCP server, not an ORM skill or static SQL linter.
FAQ
Who is io.github.YawLabs/postgres-mcp for?
Developers and small teams building on PostgreSQL who use AI agents for backend work and production debugging.
When should I use postgres-mcp?
Use it when writing migrations, optimizing queries, verifying schema, or checking database health during incidents.
How do I add io.github.YawLabs/postgres-mcp to my agent?
Install @yawlabs/postgres-mcp via npm, register the stdio MCP server, and provide a Postgres connection URL with least-privilege credentials.