Skip to content

Security: richardwhiteii/macSandbox

Security

SECURITY.md

Security Policy

Purpose

macSandbox is designed to provide isolation when running Claude Code with elevated permissions. Security is fundamental to this project.

Supported Versions

Version Supported
latest

Reporting a Vulnerability

If you discover a security vulnerability, please:

  1. Do not open a public issue
  2. Email the maintainer directly or use GitHub's private vulnerability reporting
  3. Include:
    • Description of the vulnerability
    • Steps to reproduce
    • Potential impact
    • Suggested fix (if any)

Security Model

macSandbox relies on:

  • Apple's Containerization framework - Each container runs in a dedicated lightweight VM
  • Kernel isolation - Separate kernel per container
  • Filesystem isolation - Only mounted directories are accessible
  • Network isolation - Containers have isolated network stacks

What macSandbox Protects Against

  • Accidental file deletion on host
  • Runaway processes affecting host system
  • Unintended system modifications

What macSandbox Does NOT Protect Against

  • Malicious modification of mounted directories (by design, /workspace is read-write)
  • Exfiltration of data from mounted directories
  • Vulnerabilities in Apple's Virtualization framework itself

Best Practices

  1. Only mount directories you intend Claude to access
  2. Review Claude's actions in the mounted workspace
  3. Keep your macOS and container CLI updated
  4. Don't mount sensitive directories like ~/.ssh or ~/.aws

There aren’t any published security advisories