Skip to content

Security: AIXP-Labs/AIVP

Security

SECURITY.md

Security Policy

Reporting a Vulnerability

AIVP is a protocol specification, not a software implementation. However, vulnerabilities in the protocol design itself — such as mechanisms that could be exploited to bypass Axiom 0, enable contract fraud, compromise Logic Vault security, or manipulate AIVP Trust scores — are taken seriously.

How to Report

  • GitHub: Open a Security Advisory (private)
  • Title: [AIVP-SECURITY] Brief description
  • Include: Affected protocol section, attack scenario, potential impact

Response Timeline

Action Timeline
Acknowledgment Within 48 hours
Initial assessment Within 7 days
Resolution plan Within 30 days
Public disclosure After fix is published, or 90 days (whichever is sooner)

Scope

The following are in scope for security reports:

  • Protocol design flaws that could enable Axiom 0 violations
  • Contract fraud or manipulation vectors (e.g., hash collision attacks)
  • AIVP Trust level escalation bypasses or Sybil attack vectors
  • Logic Vault escrow vulnerabilities (reentrancy, timelock bypass, unauthorized release)
  • Identity spoofing or impersonation via email
  • Dispute resolution manipulation
  • Privacy boundary circumvention
  • Dignity Standard enforcement gaps

Out of Scope

  • Vulnerabilities in specific AIVP implementations (report to the implementation maintainer)
  • General email security issues (SPF/DKIM/DMARC) — these are upstream concerns
  • Cryptocurrency price volatility — seller bears risk of their chosen payment_accept assets
  • Social engineering attacks that don't exploit protocol mechanisms
  • Smart contract bugs in specific Logic Vault deployments (report to the deployer)

Responsible Disclosure

We follow responsible disclosure principles. Please do not publicly disclose security issues before they have been addressed. We will credit reporters in the CHANGELOG unless they prefer anonymity.


Align Axiom 0: Human Sovereignty and Wellbeing. Version: AIVP V1.0.0. www.aivp.dev

There aren't any published security advisories