Read Only Local Postgres Mcp Server
Expose a local PostgreSQL database to your agent for read-only SQL so you can debug models and queries without write risk.
Overview
Read-only Local Postgres MCP Server is a MCP server for the Build phase that serves read-only PostgreSQL queries to Claude Desktop via stdio npm.
What is this MCP server?
- stdio npm package @hovecapital/read-only-postgres-mcp-server version 0.1.0
- Read-only PostgreSQL access tailored for Claude Desktop workflows
- Local instance connectivity for Docker or native Postgres dev setups
- GitHub source at hovecapital/read-only-local-postgres-mcp-server
- Sibling pattern to the Hove Capital read-only MySQL MCP server
- Package version 0.1.0
- npm identifier @hovecapital/read-only-postgres-mcp-server
- Transport stdio listed in server manifest
Community signal: 2 GitHub stars.
What problem does it solve?
Local Postgres debugging stalls when you copy SQL back and forth yet fear giving an agent credentials that can mutate dev data.
Who is it for?
Builders running local PostgreSQL who want MCP-assisted SELECT exploration during feature development.
Skip if: Running schema migrations, seed scripts with writes, or centralized production Postgres administration through this read-only server.
What do I get? / Deliverables
Agents inspect Postgres schemas and run safe read queries in-place so backend debugging moves faster with bounded database risk.
- Read-only query results and schema insights from local Postgres
- Faster ORM and migration debugging with agent participation
- Reduced accidental write risk compared with full-admin DB credentials
Recommended MCP Servers
Journey fit
Postgres is the default datastore for many indie SaaS stacks; grounding agents during Build cuts iteration time before Ship checks. Backend placement reflects schema exploration and SQL debugging on Postgres rather than distribution or monitoring setup.
How it compares
Local read-only Postgres MCP adapter, not a hosted Supabase management skill or full DBA toolchain.
Common Questions / FAQ
Who is capital.hove/read-only-local-postgres-mcp-server for?
Solo developers with local PostgreSQL who need read-only MCP SQL access from Claude Desktop while building backends.
When should I use capital.hove/read-only-local-postgres-mcp-server?
Use it during Build when validating schemas, debugging ORM queries, or explaining data relationships without granting write SQL to the agent.
How do I add capital.hove/read-only-local-postgres-mcp-server to my agent?
Install @hovecapital/read-only-postgres-mcp-server from npm, add the stdio server block to your MCP client config, and supply your local Postgres connection parameters.