We take security seriously. If you discover a security vulnerability in Zpx, please report it responsibly.
Please do NOT open a public issue for security vulnerabilities.
- Preferred: Open a private security advisory on GitHub.
- Alternative: Email the maintainers with a detailed description.
Include as much of the following as possible:
- A minimal, reproducible test case
- The Zpx and Python versions affected
- Expected vs. actual behavior
- Any impact assessment (if known)
| Version | Supported |
|---|---|
| 0.2.x | ✅ |
| < 0.2 | ❌ |
- We will acknowledge reports within 5 business days.
- We aim to triage within 10 business days.
- Fixes are released as soon as practical; public disclosure happens after a fix is available.
Zpx is an interpreted language. Like any runtime, untrusted code should be run in a sandbox. The zpx run and zpx ai commands can execute filesystem, network, and database operations — only execute Zpx programs from sources you trust.