Vaprix AI Cyber Shield
VaprixAI
CYBERSECURITY LLM
Digital Rules of Engagement (RoE) • CFAA & IT Act 2000 Safe Harbor

Target Authorization Agreement

This Target Authorization Agreement and Rules of Engagement ("RoE") constitutes a mandatory legal covenant between Vaprix AI Technologies Pvt. Ltd. and the subscriber organization, formally conferring express authorization to execute autonomous security testing against designated digital assets.

Standard Version: ROE-LEGAL-3.1•Enforcement: Mandatory Pre-Scan Verification•SHA-256 Digitally Signed

1. Ownership Warranty & Authorization Grant

By configuring, registering, or initiating a security scan on any IP address, hostname, web domain, API endpoint, or source repository, the Client represents and warrants that:

  • The Client is the sole rightful owner, operator, or duly authorized licensee of the target systems.
  • The signatory possesses explicit legal authority to grant permission for autonomous security assessments.
  • The Client has notified third-party cloud hosting providers (e.g., AWS, Azure, GCP, Cloudflare) in compliance with applicable penetration testing policies.

2. Approved Testing Policies & Scope Boundaries

All scans executed via Vaprix AI operate under strict cryptographic scope bounds and policy restraints:

SAFE Policy (Default)

Passive reconnaissance, non-intrusive banner interrogation, header inspection, zero state modification. Guaranteed zero service degradation.

CONTROLLED Policy

Active proof-of-concept injection, BOLA IDOR verification, and SQL parameterization checks with sliding-window rate throttling and safe non-destructive payloads.

3. Strictly Prohibited Actions

Prohibited Activities Under All Circumstances
  • Distributed Denial of Service (DDoS) or high-volume bandwidth saturation attacks.
  • Unauthorized access to internal cloud metadata IP ranges (`169.254.169.254`, `metadata.google.internal`).
  • Permanent modification, destruction, or deletion of client production databases.
  • Physical social engineering, employee phishing, or physical site intrusion.

4. Emergency Kill Switch & Immediate Halt

The Client maintains unconstrained right to terminate any active assessment at any millisecond. Triggering the emergency kill switch via the platform UI (`/workflow` or `/dashboard`) or API endpoint (`POST /workflow/:id/kill`) triggers an immediate `SIGTERM` halt to all active testing workers, releases locked threads, and flushes diagnostic audit logs to disk.

5. Safe Harbor & Legal Non-Prosecution

Activities conducted in accordance with this Agreement are deemed authorized access under Section 1030 of Title 18 of the United States Code (CFAA), Section 43/66 of the Information Technology Act 2000 of India, and similar global computer crime statutes. Neither party shall initiate legal action against the other for testing performed strictly within the agreed boundaries and approved policy.