Skip to content

Security: Abhi190702/Heliox-OS

Security

SECURITY.md

πŸ” Security Policy

Helix OS is a privacy-first, autonomous AI agent that executes real actions on your system. Security isn't an afterthought here β€” it's core to how we build. If you've found a vulnerability, thank you for taking the time to report it responsibly. This document explains how.


πŸ›‘οΈ Supported Versions

We actively maintain and patch the main branch only.

Version Supported
main (latest) βœ… Active
Older releases ❌ Not supported

πŸ” Scope

This policy covers the following Helix OS components:

  • 🧱 Sandbox Execution β€” isolated environments where code and plans are executed safely
  • πŸ”‘ Permission Tiers β€” the five-tier permission system with confirmation gates and rollback support
  • 🧩 Plugin Loading β€” how third-party plugins are installed, verified, and executed at startup
  • 🌐 WebSocket IPC β€” the communication bridge between the Tauri UI and Python Daemon
  • πŸ€– Python Daemon β€” the core backend driving agent orchestration, planning, and verification

Out of scope:

  • Third-party plugins not maintained by the Helix OS team
  • Vulnerabilities in upstream dependencies (please report to their maintainers)
  • Social engineering attacks

🚨 Reporting a Vulnerability

Please do NOT open a public GitHub issue for security vulnerabilities.

Public issues expose the problem to everyone before it's fixed β€” which could put users at risk. Instead:

  1. GitHub Private Advisory (preferred) β€” use the Security tab to submit a private report directly.
  2. Email the maintainer β€” contact @VyomKulshrestha via the email listed on their GitHub profile.

What to include in your report

  • A clear description of the vulnerability and where it lives
  • Steps to reproduce the issue
  • The potential impact (e.g., sandbox escape, permission bypass, plugin hijack)
  • Any suggested fixes or mitigations (totally optional, but appreciated!)

⏱️ Responsible Disclosure Timeline

Once we receive your report, here's what you can expect from us:

Timeframe Action
Within 48 hours We acknowledge your report
Within 7 days We assess severity and confirm the vulnerability
Within 30 days We develop and test a fix
Within 45 days We release a patch
After patch release Public disclosure, coordinated with you

We ask that you:

  • Give us a reasonable window to fix the issue before going public
  • Avoid exploiting the vulnerability beyond what's needed to demonstrate it
  • Not access or modify other users' data

We'll do our best to resolve it as quickly as possible.


⚠️ Severity Levels

Level Description
πŸ”΄ Critical Remote code execution, full system compromise
🟠 High Sandbox escape, privilege escalation, plugin permission bypass
🟑 Medium Information disclosure, partial access control bypass
🟒 Low Minor issues with limited real-world impact

πŸ“¬ Contact


πŸ™Œ Acknowledgements

Security researchers who responsibly disclose vulnerabilities will be credited in our release notes β€” unless you'd prefer to stay anonymous, which is completely fine.

This project participates in GSSoC 2026 and NSoC. All contributors are expected to follow this security policy when handling security-related work.

There aren't any published security advisories