Security & Trust
Security Practices & Safeguards
At MZ Nexora, security is engineered into every line of code, database schema, and AI automation pipeline we build.
Last Updated: September 2026
1. End-to-End Encryption & Secret Management
All sensitive credentials, API keys, database connection strings, and tokens are stored using encrypted environment secret managers (Vercel, Supabase Vault, AWS KMS). Environment variables are strictly isolated per deployment phase.
2. Zero-Retention AI Agent Pipelines
When integrating LLM and RAG pipelines (FastAPI / LangChain / LlamaIndex), we enforce Zero Data Retention policies where provider APIs do not train models on your proprietary business input or context vectors.
3. Self-Hosted n8n Infrastructure
Our n8n automation instances are self-hosted on isolated VPS instances protected by Docker containerization, firewall policies, SSL certificates, and IP restrictions, eliminating third-party task inspection.
4. Strict Access Control & Authentication
Client admin dashboards and LMS web portals use modern authentication standards (JWT, OAuth 2.0, Row-Level Security in PostgreSQL/Supabase) to ensure strict role-based authorization.
5. Automated Backups & Disaster Recovery
Database instances and production application states undergo automated daily snapshots and point-in-time recovery configurations to protect against unexpected data loss or hardware failures.
6. Responsible Vulnerability Disclosure
We proactively monitor dependencies via automated security audits. If you discover a potential vulnerability in any of our web apps, please report it to our security lead for immediate remediation.
Security Hotline & Inquiries
To report a security concern or request detailed architecture security reviews for your project, contact us at mznexora@gmail.com.