mcpField
O
OwnPG

OwnPG

Local
github.comDatabases0Updated 7w ago

About OwnPG

OwnPG is a PostgreSQL DBA tool for AI clients using MCP, scoped to a single database and schema. It is still in development, with no release published yet.

Key features

  1. 01Serve one PostgreSQL database over stdio
  2. 02Read-only access by default
  3. 03Select a schema for serving
  4. 04Check connectivity with ownpg doctor

Use cases

  1. 01Expose a PostgreSQL database to an AI client
  2. 02Limit database access to a chosen schema
  3. 03Check a database connection before serving

MCP support

OwnPG serves a PostgreSQL database to AI clients over MCP, using stdio. Its documented example runs read-only against one database and the default public schema.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

No install method listed yet

We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.

Repository
Packages
cargoownpg0.1.2
mcpbownpg-0.1.2-darwin-arm64.mcpb0.1.2
mcpbownpg-0.1.2-darwin-x64.mcpb0.1.2
mcpbownpg-0.1.2-linux-arm64.mcpb0.1.2
mcpbownpg-0.1.2-linux-x64.mcpb0.1.2
mcpbownpg-0.1.2-win32-arm64.mcpb0.1.2
mcpbownpg-0.1.2-win32-x64.mcpb0.1.2

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

OwnPG releases

Release artifacts, installers, and bug reports for OwnPG, the PostgreSQL DBA tool for AI clients over the Model Context Protocol. This repository is where every install path resolves and where problems are reported.

Status

OwnPG is in development. No version has been released yet. The first release appears on the Releases page of this repository when the maintainer declares the tool stable.

Prerequisites

  • A reachable PostgreSQL server, version 14 through 18.
  • Rust 1.89 or newer, only if you install with cargo install or cargo binstall.

Installing

Every path below resolves against the releases here.

  • From crates.io, building locally: cargo install ownpg --locked
  • A prebuilt binary through cargo: cargo binstall ownpg
  • The shell installer (macOS and Linux) and the PowerShell installer (Windows) are attached to every release.
  • Homebrew: brew install devops-infinity/tap/ownpg
  • npm: npm install -g @devops-infinity/ownpg

Eight prebuilt targets ship with every release: macOS on Apple Silicon and Intel, Linux with glibc and fully static with musl on x86_64 and aarch64, and Windows on x86_64 and ARM64.

Usage

Serve one database over stdio, read-only, with the default schema public:

ownpg serve -d app

Run ownpg doctor to check the connection before serving it live, and ownpg man for the full command reference.

Verifying a download

Every release carries sha256.sum with the checksum of every archive, and sha256.sum.minisig, a minisign signature over that file. The public key is published in this README at the first release. Verify with minisign -Vm sha256.sum -P <public key>, then sha256sum -c sha256.sum (or shasum -a 256 -c on macOS). Every archive also carries a CycloneDX SBOM and THIRD-PARTY.txt with the license of every dependency, and the binary embeds its dependency list for cargo audit bin.

Reporting a problem

Open an issue here with the template that fits: a bug, or a connection problem. Include the ownpg --version line and the ownpg doctor output with any password removed.

Security problems go through private vulnerability reporting on this repository, never through a public issue. See SECURITY.md.

Maintenance

OwnPG has one maintainer and one publisher. Outside code contributions are not accepted; bug reports are welcome. A successor will be named before the project is relied on by others.

License

The artifacts are licensed under either of the MIT license (LICENSE-MIT) or the Apache License, Version 2.0 (LICENSE-APACHE), at your option.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.