Skip to content

Security: briacdev/pgcopycat

Security

SECURITY.md

Security Policy

Supported Scope

pgcopycat is a local developer tool for PostgreSQL refresh and cloning workflows. Security reports are especially relevant for:

  • Credential handling
  • Plain-text config persistence
  • Unsafe destructive behavior on destination databases
  • Command execution and shell interaction
  • Unexpected leakage of sensitive data in reports or logs

Reporting a Vulnerability

Please do not open a public GitHub issue for a suspected security vulnerability.

Instead:

  1. Contact the maintainer privately through the repository contact channel or private security reporting mechanism, if available.
  2. Include a clear description of the issue, impact, affected versions, and steps to reproduce.
  3. If possible, include a minimal config or command example with secrets redacted.

If no private reporting channel is available yet, create one before widely publishing the repository or add a dedicated maintainer security email to this file.

What to Include

Helpful details:

  • pgcopycat version or commit
  • Operating system
  • PostgreSQL version
  • Command or config used, with secrets removed
  • Expected behavior
  • Actual behavior
  • Impact assessment

Sensitive Data Guidance

Because pgcopycat supports plain-text passwords for local development:

  • Never publish real credentials in issues or pull requests
  • Redact connection strings, passwords, hostnames, and dumps before sharing
  • Do not commit local config files containing credentials
  • Review generated reports before attaching them publicly

Security Philosophy

This project aims to be conservative by default:

  • explicit confirmation before destructive operations
  • identity checks between source and destination
  • local config files ignored by Git
  • early detection of missing tools and extensions

That said, this is a developer convenience tool, not a hardened secret-management product. Use it with care and prefer isolated local or non-production targets whenever possible.

There aren't any published security advisories