Security
Detailed information about Adfynx security architecture, access controls, credential protection, and security operations.
Security Architecture
Network Security
- All data transmission uses TLS 1.2 or higher encryption
- API endpoints protected by rate limiting to prevent abuse
- Production environment completely isolated from development/testing
- Database only allows connections from application servers, not exposed to public internet
Application Security
- Input validation and output encoding to prevent injection attacks
- CSRF protection and secure session management
- Passwords hashed using industry-standard algorithms (bcrypt/Argon2)
- Sensitive operations require re-authentication
Credential & Key Management
Meta Access Token Protection
This is the most critical part of our security architecture, as Access Tokens are the keys to your Meta account.
Storage Method
- Access Tokens encrypted at application layer using AES-256 before database storage
- Encryption keys stored in separate key management service (AWS KMS)
- Encryption keys rotated regularly (every 90 days)
Access Control
- Tokens only briefly decrypted in memory when calling Meta API
- Decrypted tokens never written to logs, sent to frontend, or stored on disk
- Only authorized backend services can decrypt tokens
Revocation & Rotation
- You can disconnect anytime in Adfynx platform, we immediately delete stored tokens
- You can also revoke authorization directly in Meta account settings, tokens become invalid immediately
- We support automatic token refresh when needed
Leak Prevention
- All logs and error reports automatically sanitized before recording, removing tokens, keys, and other sensitive info
- Development and testing environments use mock data, no access to production tokens
- Code review process ensures no accidental commits of sensitive information to version control
Internal Access Control
Least Privilege Principle
We strictly limit who can access production data and systems:
Role Separation:
- Developers: Can access development and testing environments, cannot directly access production database
- Operations Engineers: Can access production infrastructure, but require two-factor authentication (2FA)
- Support Staff: Only access necessary account information through restricted admin interface, cannot access raw database
Production Access Audit
- All production environment access is logged (who, when, what they did)
- Audit logs are immutable and regularly reviewed
- Abnormal access patterns trigger automatic alerts
Employee Security
- All employee accounts require two-factor authentication (2FA)
- Regular security awareness training
- Departing employee access revoked on departure date
Vulnerability Management
Dependency Scanning
- Automated tools scan third-party dependencies for known vulnerabilities
- Critical vulnerabilities fixed within 24 hours, medium within 3 days
Code Review
- All code changes require review by at least one other engineer
- Security-related changes require additional security review
Security Monitoring & Response
Real-time Monitoring
- Application performance and error monitoring
- Database query monitoring and anomaly detection
- Login failures, abnormal API calls, and other security event alerts
Incident Response Process
- Detection: Automatic alerts or user reports
- Assessment: Determine incident nature and impact scope (Sev1/Sev2/Sev3)
- Containment: Isolate affected systems, prevent further damage
- Eradication: Fix vulnerabilities or remove threats
- Recovery: Restore normal service
- Post-mortem: Analyze root cause, improve processes
Customer Notification
- Sev1 (data breach, complete service outage): Notify within 24 hours
- Sev2 (partial functionality affected): Notify within 3 business days
- Sev3 (minor impact): Explain in next regular update
Data Backup & Recovery
Backup Strategy
- Daily automatic database backups
- Backup data encrypted in storage
- Backup retention period: 30 days
Disaster Recovery
- Recovery Time Objective (RTO): 4 hours for critical services
- Recovery Point Objective (RPO): 24 hours maximum data loss
- Recovery process tested quarterly to ensure effectiveness
- Multi-region deployment capability for enhanced resilience
Verifiable Statements
- We provide account activity logs; you can view recent login records and API calls in account settings
- We log all internal production environment access; audit logs retained for 12 months
- We support revoking Meta connection anytime; tokens immediately deleted from our system
- Our infrastructure providers (Vercel, Railway, Supabase) all provide DDoS protection and auto-scaling
- We use AES-256 encryption to store Meta Access Tokens; keys managed separately
Frequently Asked Questions
A: We implement least privilege principles and role separation. Production data access requires two-factor authentication (2FA), all access is logged and audited. We regularly review access logs, and abnormal behavior triggers alerts.
Last updated: December 19, 2025