Security Overview
Your prompts, evaluation data, and production traces are core IP. We built Agenta to protect them with the same rigor you'd expect from any critical infrastructure: SOC 2 Type II certified, AES-256 encryption at rest, and data residency options in both US and EU regions.
Trust Center
Review our security controls, compliance certifications, and policies in our Trust Center.
Compliance
Agenta is SOC 2 Type II certified. For details and to request access, see our Trust Center.
Encryption
Data at Rest
All sensitive data stored in Agenta is encrypted at rest using AES-256 encryption:
- Secrets and API keys: Encrypted using PostgreSQL's PGP symmetric encryption (AES-256)
- Databases and storage: Encrypted at rest
Data in Transit
All data transmitted to and from Agenta is encrypted using TLS 1.2+:
- HTTPS enforced on all endpoints
- No plaintext data transmission
Infrastructure Security
Cloud Infrastructure
Agenta Cloud runs on AWS with SOC 2 compliant infrastructure:
- Network isolation: Dedicated VPCs with public/private subnet isolation
- High availability: Automatic failover across availability zones
- Redundant networking: Built-in redundancy for reliable connectivity
- Security groups: Principle of least privilege access rules
- DDoS protection: Included by default
Access Controls
- Role-based access control (RBAC): Granular permissions for organization members
- SSO integration: Single sign-on via OIDC (Business and Enterprise)
- API key management: Scoped API keys with rotation support
- Audit logging: Track access and changes
Application Security
Authentication
- Multiple authentication methods: email:password/OTP, social OAuth, SSO
- Session management: Secure session handling with automatic expiration
- Account linking: Safely link multiple authentication methods
Authorization
- Organization isolation: Data isolated between organizations
- Permission checks: Every API request validated against user permissions
- Resource scoping: API keys scoped to your organization
Monitoring & Incident Response
System Monitoring
- Status page: Current status and incident updates at status.agenta.ai
Incident Response
- We publish updates and incident timelines on the status page
Reporting Security Issues
If you discover a security vulnerability, follow our coordinated disclosure process:
- Security policy: https://github.com/Agenta-AI/agenta/security/policy
- Email: security@agenta.ai