Security and trust
Who sees what, and how you prove it.
Shipeezi handles the data a retail import chain depends on: prices, origins, identities and the trace itself. The model below is designed so an LSP can operate for many clients without a boundary ever leaking.
The model
Identity, isolation and the audit trail.
Identity
One identity provider across the product, with an audience per API. Access groups bundle permissions and users inherit them, so a role is a bundle, not a per-screen grant.
Tenant isolation
Managed accounts partition per client by grant. Every query is scoped by both the operating entity and the client, so an LSP never sees across a boundary it was not granted.
Audit
A generic change-event log with an interceptor records who changed what, when. It is the same trail that carries PO change events and any action the platform applies on its own.
Data handling
Sensitive trade and personal fields use field-level encryption. Retention and erasure are policy-driven per record type, not hardcoded.
Agent governance
Anything applied automatically runs at a set level per capability, is logged, and is reversible. There is an explicit agent-action audit trail.
Login policy
Lockout ladder, idle termination with a warning, minimum complexity and password-reuse rules, matched to the incumbent product's thresholds for parity.
Trust documents
The paperwork, when you need it.
A security pack for your review.
Ask us for the hosting, data-residency and sub-processor detail your procurement needs.