Security & trust

Your customer data is safe.

Agentez handles every customer your business has — names, phone numbers, addresses, payment links, photos. We treat that data the way we'd want our own treated.

lock

Encrypted in transit & at rest

TLS 1.3 in transit. AES-256 at rest. Database backups encrypted. Zero plaintext storage of credentials.

verified_user

Row-level security

Every owner-scoped table is RLS-protected at the database layer. Two operators can't see each other's data even with a leaked API token.

cloud_sync

SOC 2 Type II in progress

Working with Vanta toward Type II certification. Quarterly penetration tests by an independent third party.

key

SSO & 2FA

Sign in with Google, Apple, or Microsoft. Two-factor authentication available on every plan.

history

Audit log

Every action — yours and the agents' — is logged with a timestamp, source, and outcome. Exportable for any reason.

delete_forever

Right to delete

Cancel and we delete everything within 90 days. Or hit "delete now" and we delete within 24 hours. No retention games.

AI specifically

How we handle AI and your data.

Your prompts and your business data are never used to train any model — ours, Anthropic's, OpenAI's, Google's, or anyone else's. We pay for inference. Period.

Where AI runs:

  • Anthropic Claude (Haiku) for most agent tasks. Anthropic's API has zero-retention enabled by default for our account.
  • OpenAI for specific cases that need it. Same zero-retention setup.
  • Google Vision and Imagen optionally for photo and image work.

No customer data is stored on a model provider's servers beyond the request lifetime. We log only what's required for debugging, and that log is purged within 30 days.

Found a vulnerability? Email security@agentez.com. We pay bounties from $250 (informational) to $10,000+ (critical). Responsible disclosure honored.

Built so you can hand it your business.