Skip to content

Security: borhen68/phantom

Security

SECURITY.md

Security Policy

Scope

PHANTOM is an autonomous agent runtime with tool execution, persistent memory, and generated skills. Treat it as a sensitive local automation system.

Reporting

If you find a security issue, please report it privately before opening a public issue.

Until a dedicated security inbox exists, use GitHub private reporting if available for the repository.

High-Risk Areas

  • generated skill execution
  • shell and file tool safety boundaries
  • messaging webhook authentication
  • secret loading and redaction
  • browser automation against privileged systems

Current Security Posture

PHANTOM includes:

  • scoped local state
  • tool confirmation controls
  • webhook signature verification
  • provider secret redaction
  • allowlist-based skill validation
  • Linux sandbox preference for generated skills

PHANTOM does not yet provide full hostile-code isolation. Do not treat it as a hardened multi-tenant execution platform.

There aren’t any published security advisories